Conference 7.286::home_work

    I bought a GFI outlet to install in an outdoor outlet box.
    
    Question:  Could I install the device as the outlet of an extension
    wire and then use the extension plugged into a two-prong (non-grounded)
    outlet?   Would it offer me any protection?
             
    
    I just bought a microwave oven for use in a cottage which doesn't have
    grounded outlets. 
    
    Question:  Should I not use it without adding 3-prong sockets and 
    upgrading the wiring to include grounding?
    
    Dave
    
    

>    I bought a GFI outlet to install in an outdoor outlet box.
>    
>    Question:  Could I install the device as the outlet of an extension
>    wire and then use the extension plugged into a two-prong (non-grounded)
>    outlet?   Would it offer me any protection?

No and no:

This sort of mixture of permanent and temporary wiring is not permitted by the
National Electrical Code.  Components of permanently installed wiring (e.g.
outlet boxes, wiring inside walls) and of temporary wiring (e.g. extension
cords) are used for different purposes, have different requirements for such
things as rigidity vs. flexibility, are manufactured to different standards,
etc.  A GFI outlet is permanent wiring, and shouldn't be installed in a
temporary manner such as you describe. 

A GFI is useless without a ground anyway.
             
    
>    I just bought a microwave oven for use in a cottage which doesn't have
>    grounded outlets. 
>    
>    Question:  Should I not use it without adding 3-prong sockets and 
>    upgrading the wiring to include grounding?
    
Strictly speaking, you should indeed not use it without adding grounded service.

However, you're presumably already using electrical appliances in this cottage 
with the electrical service the way it is.  I don't know of anything special 
about microwave ovens that make them more critical to ground properly than
other types of kitchen appliances.  So using the oven ungrounded probably
doesn't make your kitchen any more dangerous than it already is. 

Keep in mind that internal insulation failure in any electrical appliance can
cause exposed metal parts of the appliance to become electrically charged,
often dangerously so.  If your service is grounded, this condition causes a
breaker to trip or a fuse to blow right away.  If your service isn't grounded,
you find out about the condition when you get a shock from touching those
metal parts.  As grounded service has become common, some appliance
manufacturers have let the quality of their internal insulation slide. 

An alternative to re-wiring the entire cottage is just re-wiring the kitchen.  

An even simpler workaround is to use a 3-prong-to-2-prong "cheater" adapter, 
and run a wire from the adapter's grounding lug to a convenient ground such as 
a water pipe.

    Another note, are you sure that you have no ground at the outlets?
    Alot of older houses use BX (metal sheathed) wire and only used
    two-prong outlets.  The outlet box is grounded because it touches
    the metal sheath.  You may just have old outlets.  Don't just replace
    them with new ones until you check it out.
    
    mac

    Re .28 & .29
    
    The reason I asked about putting a GFI in an extension wire is that
    quite frequently I have to use portable power tools at the cottage,
    and sometimes right down at the waterfront.  I was under the impression
    that a GFI operates by sensing the difference between the incoming
    current and the current returning on the neutral line.  If there
    is a difference, the circuit breaks.  If that's the case, then what
    is the additional ground needed for?  Besides, I had no intention of
    trying to defeat the purpose of the electrical codes.  As I asked before,
    are my fears "groundless" (pardon the pun) ? 
           
    Dave

    I don't know where you live or if it helps at all, but for what
    it's worth, Spag's has extension cords with GFI outlets on the end.
    Perhaps some other store near you carries them, too.

    I have recently seen an extension cord/GFI outlet combo. Its a heavy
    duty extension cord with GFI protected outlets molded in plastic at one
    end. It's ideal for protecting your self while doing those outside
    chores. 

    As stated earlier though, it has a grounded plug that shouldn't be
    compromised. 
    
    Charly

Re .29: It would take me all night to find it in the code, but as I
recall, BX sheathing is not considered a suitable ground. At least,
I was brought up to consider this a no-no. It has too high a
resistance, and the connection between the sheath and the box may
corrode and get flaky.

    re: .33
    As I understand it, there's a metal strip inside the BX to provide 
    a good ground path.  You should see it when you cut the BX.  I think
    you bend the strip over the end of the BX so it gets clamped by
    the cable clamp on the box, and that provides the ground.  (My
    knowledge of how to work with BX is minimal at best, so anybody
    else who knows better please correct me.)

    re: -1
    		You are correct.  There is a thin metal strip (not a
    wire, its flat) inside the metal sheath.  After the BX has been
    cut, and the plastic insulator has been put in (my boss called them
    "red caps") the metal strip is wrapped on the outside of the BX
    a distance long enough so that it will contact the metal box you
    are bringing the wire into.  I would like to know if it is not a
    "valid" ground, but I do think it is.  
    
    mac  who_hates_to_work_with_BX_but_will_take_money_any_way_he_can_get_it

    FYI, the electrical code makes no mention of BX at all, making me
    think it isn't "supported" any mor-reed

    try looking for "armored cable" in the code.  I'm 99% sure the code
    still allows it, as it's discussed in all three of my DIY electrical
    books (I don't need them, I just like looking at the pictures! - (:> )
    
    Jim

    I know it is still allowed.  Go look at any new construction where
    steel 2 by 4's are being used.  Every time I have seen these walls
    there has been BX (armored cable) used in them.
    
    mac

    RE: -3?
    
    	Older BX (I am not sure when they started) does not have the
    flat strip inside.
    
    mac

OK, I stand corrected. This evening I was awake enough to dig through
my 1984 electrical code. 250-91(b) says:

"Types of equipment grounding conductors. The equipment grounding conductor
run with or enclosing the dircuit conductors shall be one or more or a
combination of the following: (1) a copper or other corrosion-resistant
conductor .... (6) armor of Type AC cable ..."

AC is the generic code for so-called BX. (Got that?)

    BX would not be in the code as it is a trademark (you don't find
    Romex either).  

    OKOKOK!  I never realized BX ws the trademark; armored cable I know
    about.  The code's description (section 333-4) "Cables of the AC
    type, expect ACL [lead covered conductors for areas exposed to
    weather], what have an internal bonding strip of copper or aluminum,
    in intimate contact with the armor for its entire length."  Nothing
    in section 333-9 on boxes and fittings mentions the internal strap
    - just that you have to have a fitting that clamps to the armor
    for grounding.  AC is legal for branch circuits and feeders, for
    both exposed and concealed work, and can be fished through walls,
    etc., except in damp locations (where type ACL must be used).  OK
    to run AC through the air voids in concrete blocks or tile walls.
     Cannot bury AC in the earth - must be in a conduit.  In attics
    with non-permenant stairs for access, it only needs to be protected
    with guard strips for the area within 6' of the entrance/scuttle
    hole; otherwise it has to have the guards along the entire length.
     It can be run along the exposed underside of floor joists in basements
    if not subject to physical damage.
    -reed

    Just to add to the armored cable trivia here, there is also a variant
    of wiring in which wires are inserted into flexible conduit.  This
    conduit looks just like the armor of type AC.
    
    In the tradition of BX and Romex, this is also commonly known under
    a tradename.  This name escapes me at the moment but I think it
    begins with "Green.." something.
    
    Rich

RE:1204.44 
    
>    In the tradition of BX and Romex, this is also commonly known under
>    a tradename.  This name escapes me at the moment but I think it
>    begins with "Green.." something.
    
	Greenfeld or Greenfield tubing is just the flexible armor sheathing
	into which you snake the individual wires.  Often used for the feeder
	from junction box to recessed light.

    RE:.44
    	When I was working as an apprentice, this stuff was refered
    to as "FLEX".  We used to make the connections between the switches
    on the side of gas furnaces to the motor.  There is a code ruling
    that says something about no-plastic sheathed cable within some
    distance of a furnace.  The FLEX made things easy.
    
    	The FLEX (again probably a brand name) is alot more flexible
    then the armored cable and it comes hollow (no wire).
    
    mac

I was told that any electrical outlets in wet places (bathrooms, kitchens)
should have "Ground Fault Circuit Interrupters" on them. 

For the record, could someone explain exactly what they are and how to replace
an old-style outlet with a GFCI? 


- Brian

    layman's explaination: they detect that all the current that is
    coming in over the "hot" line is not going out over the "common"
    line and it then throws its own local circuit breaker.  If the 
    "lost" current is about 5ma(?) or more it assumes that either you or
    something else is conducting the "lost" current.
    
    If by "old style" you mean a two prog outlet, you'll have to supply
    a good ground wire; otherwise I believe you just replace the old
    outlet.
    
    rmm
    

	In simple terms, the GFCI monitors the current passing through
	the hot (black) and neutral (white) wires downline of the unit.
	If there is parity, that is the same amount of current going
	through the hot wire is the same as the amount of current coming
	back through the neutral wire, nothing happens.

	However, if there is a difference between the two of ~40 ma,
	the unit trips. This could occur if a razor were to fall in a sink
	or if you were to use a defective power tool (hot tied to metal)
	on wet grass (good gound). Some of the current would return through
	ground (water and drain pipe with the razor example, you in the tool
	example) rather than neutral. A circuit breaker would only trip if
	it's rated capacity were exceeded. Probably too late.

	GFCI outlets also have a "pass through" feature. This allows outlets
	past the GFCI to be protected also. Hookup instructions are provided
	with the unit. It is not much more difficult to install than an
	ordinary outlet.

Larry

    
    
    I've seen where an electrician placed in a run the gfi in  
    the kitchen and standard sockets in both baths in effect 
    giving the protection to the three areas with a third the 
    cost. Just a comment............ 
    
    

    	RE .3
    		This is a good thing to think of when doing a new wire
    job. I just finished the rough in on a sunroom and new deck and
    planned for a single GFI outlet to take care of all the outlets
    in the room and on the deck. Its just a matter of making the first
    place your power FEED goes, the GFI recptacle. After that every
    thing down line from it is protected.
    
    
    				My 3� worth...
    
    						...Dave
    

As has already been pointed out elsewhere in this conference, I think:

Don't go overboard with many outlets downstream from a single GFI.  Remember, 
you have to find the GFI to reset it when it trips.  If it's many steps (or
even floors) away from the outlet that made it trip, finding out what went
wrong could become extremely inconvenient, even dangerous if lighting or
other safety gear is involved.

    Some GFCI outlets are packaged with a bunch of "Ground Fault Protected
    Outlet" stickers that you can place on downstream outlets.  This
    is useful for 1) extra piece of mind when plugging something in
    them, and 2) tracking down the problem when one of the outlets seems
    to be inoperative.
    
    __Rich

I finished a basement playroom for my kids - put in ONLY GFCI-protected
outlets since GFCI's provide some protection from electrocution - (but they
aren't 'perfect' since you can electrocute yourself across the hot and neutral
without creating a ground fault). I figured the most likely case is
metal-object-poked-into-hot-side-of-outlet which will trip a GFCI.

A couple of 'practical' notes:

1. GFCI outlets are large (physically) - so cut excess wire lengths inside
the circuit boxes to a minimum to avoid 'cramming'.

2. The special cover plates are generally flimsy - tighten screws carefully.

3. If you do replace an outlet in a line with a GFCI, make SURE it's the
FIRST one in line. The only 'REAL' way to be sure in a house you didn't
wire is: Turn power to the line off; disconnect the rest of the line of
outlets from the outlet you plan to replace; Turn ALL OTHER CIRCUITS IN
THE PANEL off; turn the line you're working on back on: Now NO outlet except
the ONE you plan to replace should have power (a lamp makes a good tester).

Dennis

	The reply I made stating that the GFCI will trip if it detects
	a 40ma difference between hot and neutral should have read 5ma.
	The time to trip is ~40ms.

Larry

    re: .3 and others
    
    If you are so cheap as to do this, tell the next owner of your house.
    It is a royal pain in the a** to run around a 60 deg (I'm cheap
    too) house in your nightie to find where the last owner (cheap bugger)
    put the GFI receptacle when your electric razor trips the damn thing
    in the morning before work.

    
    re: .9
    
    Your scenerio should only happen once, and, in your case, perhaps 
    twice if you have two baths in your house.   
    

    re: .10
    
    It is still cold even if I know where it is.  I put a GFI in the
    master bathroom.
    
    Stan

    In my house I have a GFI circuit breaker, so everything on the
    circuit is protected. Are there any pros/cons as to which installation
    might be better. 1) A GFI outlet in the first outlet box in a loop,
    or 2) a GFI in the pannel.
    
    I don't think that cost is much of a factor.
    
    

re .12, outlet GFI vs. breaker GFI

This question has been touched on in several topics in this conference.  The 
consensus, as I remember it:

1. Convenience/safety:  when a GFI trips, you need to be able to get to it, 
   both to find out that it has indeed tripped and to reset it.  If it's in 
   the breaker panel, it's pretty much guaranteed to be inconveniently 
   located.  That inconvenience becomes a safety problem if your path to 
   the breaker box is normally lit by lights on the tripped circuit.

2. GFIs packaged as breakers seem to be less reliable than ones packaged as 
   outlets.

3. Cost is indeed a factor - the breaker-packaged GFIs are more expensive.

#2 and #3 may change as the technology matures.  But it appears as though
the outlet-packaged GFIs are more popular, and hence they're receiving the
attention, technology improvements, and cost reductions. 

another point:

You use a GFCI breaker when you want to protect something other than
outlets... for example, a jacuzzi.

The breaker is usually less reliable because there is more wire for the
GFCI to detect faults in. If you use an outlet, you don't protect
the circuit from the panel to the first GFCI outlet, thus making
the GFCI circuitry more reliable for the wire it does protect.

I was installing a new GFCI-protected outlet in my bathroom, taking advantage
of the "downline" protection feature mentioned in previous notes.  The main
device I wanted to protect was the shower exhaust fan, but the overhead
light is also on the same branch circuit.

When I went to test out my new wiring job and flipped on the breaker in the
breaker box, I heard a small "pop," and groaned when I realized I must have
goofed up my wiring somewhere.  I flipped the breaker pole to off again, and
went to check the GFCI.  The RESET button was still pushed in, so I wasn't
able to reset it.  Pushing the TEST button produced no results either. 

According to the instructions that came with it (Leviton makes it), when
this happens, I'm supposed to "CALL A QUALIFIED ELECTRICIAN!" (emphasis
theirs).

I refuse to give up so easily.  Assuming I can figure where I goofed in
my wiring, can anyone explain what has happened to the GFCI?

Thanks,

Brian

      I'm not sure I understand why you expected the GFCI to trip if
    there was in fact a problem with the wiring.  If the fault is after
    the GFCI, then I would expect it to trip but if the fault was with
    wiring the GFCI or before it, I wouldn't expect it to trip.  I don't
    know if a short would trip the GFCI because it is measuring the difference
    between hot and neutral and with a short between hot and neutral
    they would still be the same.  As far as the reset button is
    concerned, this only works when power is applied...at least with 
    the Leviton's.

    -Jim

    Have someone turn on the breaker while you watch the GFI. You might
    be able to spot the problem. GFI's, being large, are you sure it
    isn't touching the side of the box?

    (Since I can't find a better place for it, I guess I'll stick this
    question in here.)
    
    I was over a friend's house last night, installing a ceiling fixture
    in the bathroom to some wiring that someone else had previously run.
    A GFCI-protected outlet was nearby, apparently installed downstream
    from the light.  After the light was installed and power turned
    back on, we noticed that the GFCI would sometimes trip just as the
    light switch was turned off or on.  When this happened, the ceiling
    light would still operate (thus, I concluded it was upstream from
    the GFCI, which therefore could not be protecting the light/switch
    wiring).
    
    As I thought about what could be causing this behavior, I remembered
    reading that GFCIs seem to go bad fairly often.  I wondered if maybe
    it was a little soft, and some voltage fluctuation as the switch was
    thrown could be tripping it.  Does this sound reasonable?  Since no
    movement of wires is involved, I find it hard to believe a loose
    wire rubbing against something it shouldn't be is at fault.  The
    lights and outlets in this bathroom have never been used, as the
    owners were using a light placed on the sink and an extension cord
    run across the floor from the bedroom (that's another story!).
    
    As a next step, I suggested replacing the GFCI.  Any comments or
    different suggestions would be appreciated.
    
    Jim

RE. -1

I can't quite visualize your set-up.  You say the GFCI is "downstream" of the
light and yet it is not "switched" by the light switch (the GFCI still has
power when the light is off, right?).  

So how does the GFCI get its power feed? 

B.

    Yes, the GFCI does still have power when the light is switched off.
    Perhaps "downstream" was a poor choice of words.  I didn't do the
    wiring, so I can only guess.  Here's a picture:
    
  power  ]--------white----+---------------------------------+
  source ]------black----+---------------------------------+ |
                         | |                               | |
                         | |                            +--------+
       +--------+    +---------+    +--------+          |  GFCI  |
       | 3-way  |====|  light  |====| 3-way  |          | outlet |
       | switch |----| fixture |----| switch |          +--------+
       +--------+    +=========+    +--------+             | |
                                                           | |
                                                           v v
    The power source is probably a feed from
    other outlets, lights, etc.                          feed to
                                                        additional
                                                         outlets
    
    From the above, I can't see how a fault in the wiring for the light
    would lead the GFCI to trip.  Also, why should it only trip some
    of the time when one of the light switches is moved?  If it's a
    fault, I would think it would trip all the time, given that wires
    are not being moved about.
    
    Jim

      Was this by any chance a fluorescent light?
    
    -Jim
    

    Nope, it's a regular 60-w incandescent light bulb.
    
    Jim

Sounds to me like there's something wrong with the continuity of the ground 
as it makes its way through those upstream switch boxes, so the GFCI is 
rightly tripping.

Although the diagram in .20 shows a splice in the supply wires themselves,
separate from the light and switches, the splice is of course actually
inside one of the boxes, probably one of the switch boxes.  So check the
switch boxes to see which one has a splice in it, and look for a ground
continuity problem there.

    Re: .23
    
      I'm not sure I understand what you are saying about the ground
    continuity causing the GFCI to trip.  It is my understanding that
    the GFCI does not use the ground but monitors the difference between
    hot and neutral and if the difference reaches its threshold, it
    trips.  This made sense to me because the code* specifies that if
    you don't have a ground when replacing a two prong receptacle, you
    must either replace it with another two prong receptacle or a GFCI
    receptacle.  Also, I had a problem a while back in which the ground
    in one of my circuits had approx 50 volts in it and yet the GFCI didn't
    trip so this also reinforces my understandings above.  What had happened 
    was that someone had spliced a piece of wire in the circuit that had no
    ground and the voltage was being induced into the ground. I suspect
    in this case that there may be a problem with the neutral.  Possibly, 
    one of the outlets is using ground instead of neutral and is causing
    the trip.
    
    *Code in this context refers to 'Wiring Simplified' by Richter and
     Schwan

    -Jim

    Why the question abour florescent lights?  I have trouble with a GFCI
    circuit breaker that is in my power panel.  This GFCI triggers fairly
    often in the summer, but never in the winter.  There is a florescent
    light on this circuit.  The triggering doesn't seem to be tied to
    anything that I've noticed.  Any ideas?
    
                                    August G. Reinig

Maybe there is dampness in the wiring somewhere causing a weak trickle
to ground, that every now and then reaches the critical point and trips it.
In the winter the air would be drier so maybe that's below critical.

Or maybe the thing overheats in some fashion?  Doesn't seem likely.

I can't think what flourescent lights would have to do with it.

    I had a problem with a GFCI and a flourescent light in my bathroom, the
    GFCI would trip when I turned on the light. The electrical supply house
    refused to believe that it was a problem with the particular GFCI. I
    switched to a different GFCI brand (Leviton), and the problem went
    away. 

    Re: .25
    
      I asked the question about the fluorescent light because of a
    situation such as yours that was mentioned to me. From what I
    remember, it was a GFCI breaker that was tripping.  The person had 
    checked out the entire circuit but could find nothing wrong.  He 
    finally disconnected the fluorescent light that was on the circuit 
    and the problem went away. I do not know why this would happen with 
    a fluorescent light, just that one was determined to be the cause of 
    this problem.

    -Jim

    As an aside...  I just replaced a GFCI outlet in a bathroom.  After
    seeing prices range from $17-$25 for these things (Leviton brand),
    I got it at Spags for $8.50.
    
    Marc

    Re: .29
    
      Note that the $8.50 price is their sale price.  I believe they
    will be at this price until the end of the week.  Their normal price 
    for GFCI's is around $13.50 but they do go on sale often.

    -Jim

A properly wired non-defective flourescent light should not trigger an equally
properly installed non-defective GFCI.

I'd suspect ground leakage in the flourescent ballast. Since the Ballast
is bolted to sheet-metal which is wired to safety ground, a leak there would
do the trick.

Dennis

    Not having much experience with electrical stuff...I will try
    to explain a problem I am having with an outside GFCI.
    
    Yesterday I plugged in the weedeater and -- nothing -- so, I
    thought, hmmmm...circuit must have tripped....reset the gfci...
    nothing happened.  So, tried resetting the breaker....nothing
    happened.  Now, this particular gfci is on the same circuit
    as two inside lamp outlets, and they work okay.  So, did 
    the gfci go bad, or is there a "leak" somewhere?  Please say
    it's the outlet, cuz all the wiring is in an essentially
    non-accessible crawlspace!!
    
    Thanx for any help on how to troublshoot this stupid thing!
    
    Deb
    

Are you sure the weedeater works properly?

Has this outlet ever worked properly?

Many GFCIs have pilot lights - does yours?  Is it lit?

Is the GFCI unit right there in the outdoor box, or does the outdoor box
contain a regular outlet, protected by a GFCI somewhere upstream? 

Are there any more outlets downstream of this one, or is it the end of the 
line?

The answers to the above questions should suggest things to try.  Note that 
wiring failures usually occur inside boxes (connections and devices); cables 
seldom fail. 

    
    Well, to answer those questions...which may shed more light on
    the situation..
    
    Yes, the weedeater works...I used another outside outlet after 
    this gfci failed.
    No, it doesnt have a "pilot light" --- and it is installed right
    in the box.
    I assume its the last thing on the circuit, as the other outlets
    work okay.
    
    I have a feeling that somehow when the folks next door were
    building their house, and "borrowed" my electricity at this
    box, they used some saw, or whatever too big, and "blew"
    the gfci itself.  Is that possible???  I do not know if it
    has worked since then, as yesterday was the first time I 
    tried it.
    
    Pushing in the test button does not cause the reset to pop
    as it does on the gfci in my bathroom....so, something
    somewhere is wrong, and I will replace the gfci, if I can't
    decide anything could be the problem.
    
    Any other possibilities???
    
    Thanx
    Deb
    

    
    	Just a thought.
    
    	GFI's are very sensitive to moisture.  I don't know where you
    live, but if it's in New England, we have had quite a few days of
    rain here lately, which may be inhibiting it's functionality.  I
    would wait till you get a good dry day (if this situation applies)
    and then try it again.

    
    Yes, I do live here in beautiful "mud england" --- 
    sheesh, have you ever seen so much rain!!  I dug up
    flowers last night which had rotted out at the roots due
    to the excessive moisture.
    In any case, I will wait till things dry up (soon hopefully)
    before replacing the gfci.
    Thanx for the tip.
    
    Deb
    

If you open up the box, you can determine with a meter or test lamp whether 
power is getting to the box at all.  If so, the problem is in the GFCI.  If 
not, it's in the wiring someplace upstream.

Being basically lazy, I would wait until things dry out.  (Safer to open up 
the box then anyway).  If I became impatient, I might help things along 
with a hair dryer or heat gun.

I've had a lot of trouble with GFCIs exposed to the elements (i.e., not
strictly indoors).  Things seem to work a lot better when an
outdoor or breezeway outlet is wired to the load side of an indoor GFCI. 
Maybe the problem has been moisture, but then, in New England that 
means your GFCI will be out of service a lot of the time. ;-)

They now sell (or you can easily make) a portable GFCI outlet.  You plug
the thing into a standard outlet and then your extension cord into the
portable outlet. Your extension cord is now GFCI protected.  You can 
store the portable GFCI outlet indoors so that it will be dry and in 
service when you need it. 

Alex

    I have another GFCI problem. I recently finished my second floor
    and installed one of these in the new bathroom. This was about three
    months ago and it was working fine. The other day I went to test
    it and it won't trip. There is electricity in it and in the other
    outlet down the line. What could be causing this? Is it time to
    get a new one? I would expect it to last longer than three months.
    After all, the one downstairs is three years old and works fine.
    With so many GFCI problems on this note I am starting to wonder
    if this is some kind of a communist plot :-). Any ideas will be
    appreciated.

    RE: .40
    
    	As I have been led to believe, GFCIs have a rather limited life
    	expectancy, usually a couple of years.  Unfortunately, this
    	also seems to include the shelf/storage time.  So, if you got
    	one which had been sitting around for a while it may not last
    	very long in service.
    
    I discovered this when I had a couple of GFI breakers go bad after
    about a month after I installed them.  I took them back (to Ralph
    Pill) because they had a year guarantee.  Well it turns out that
    the guarantee is for a year from manufacture, not from point of
    sale.  Maybe this is the manufacturer or simply the nature of the
    beast; I'm not sure.
    
    Anyway...No, I wouldn't be surprised to see a GFI outlet go after
    only a couple of months in service.
    
    - Mark

    
    
    Does anyone know how much current the downstream side of
    a GFCI receptacle is rated for?
    
    Thanks
    Len DiPinto
    

    I had the GFI in my bathroom go after two months also. It provides
    electricity but won't trip when you push the test button. It happened
    right after I used an electric drill in the outlet. My brand was
    a Levitron. Maybe if we post all the brands that fail, people can
    avoid these when installing new ones.
    
    Tim
    

>< Note 1659.42 by AD::DIPINTO >
>                           -< What is it rated for? >-
>    
>    Does anyone know how much current the downstream side of
>    a GFCI receptacle is rated for?
>    

    The ones  I  just  bought  (Leviton)  are  rated at 15 amps to the
    outlet  and 20 amp total. That is if you install them as the first
    outlet in a 20 amp circuit, each duplex outlet is good for 15 amps
    up  to  a  total  of  20  amps. Presumably all 20 amps could be in
    downstream outlets.

--David Wittenberg

>                                                    It happened
>    right after I used an electric drill in the outlet. My brand was
>    a Levitron. 

I had a Leviton croak shortly after a contractor plugged a sawzall into 
it.  I croaked another when I had a few blade stalls with a circular
saw. I think they might not do well under that kind of load, because
there are two other GFIs in the house from the same lot that that have
never been used for power tools, and they are working just fine. 

					>>>==>PStJTT

    Mine is also a Leviton. I have plugged my drill on it a few times.
    Could it be a problem with the brand not being able to handle power
    tools?

I've got a problem with my GFI breaker... this is the story...

During the Winter, I noticed that my exterior garage light would go out
if I turned it on during a heavy snow.  Each time it went out, my GFI
breaker had tripped.

I managed to put up with the problem, because I thought that I knew what
the problem was, i.e. somewhere, somehow a wee bit of snow was getting
into an exterior socket and tripping the breaker.

The problem seemed to go away in the Spring, or maybe I just didn't use
my exterior lighting all that much.  But, anyway, its now the Summer
and I'm back to base 1.

I've explored deeper into the problem.  I have confirmed that if I plug
anything at all into any of my exterior plugs, no matter what time of
the day or weather conditions, the GFI immediately trips.

I'm also reasonably convinced, that if I unplug certain "heavy" juice
items in the house, like a refrigerator for instance, the GFI can also
trip, yet these items are not necessarily on the GFI circuit.

Both of my bathrooms, my garage lights and sockets, my exterior lights,
and all of my exterior plugs are wired to the GFI breaker.  I believe
that this is normal to wire so many circuits to one GFI.  Is this true?

My suspicions are as follows{most suspicious first}:

	a) The GFI breaker itself is bad,

	b) Some part of the wiring into the GFI is faulty,

	c) The GFI is overloaded and therefore trips when anything
	   at all is plugged into an exterior socket.

Can anybody lend there suggestions or confirm the above?  Has anyone
experienced a similar problem?  How'd you fix it?

Thx. rep

    I have also had a similar problem at my house.  Mine though happens
    only during damp or muggy weather.  I have talked to an electrician
    about it and I was informed that that is normal.  Since yours is
    not limited to the weather conditions, it sounds like it should
    be replaced.
    
    Bob O.
    

Other input from around the office indicates that the GFI should be loaded
with at most 4 outlets.  I'm beating that without a doubt.

Its still somewhat of a mystery to me why the breaker trips when I plug
something in on an outside receptacle.  Maybe a little ground bounce?

Currently, it looks like the best suggestion is to split the circuit with
two separate GFI breakers, one for exterior and one for interior.  How
does that sound?

Now to see if I have anymore room in the box...

rep

>    Other input from around the office indicates that the GFI should be
>    loaded with at most 4 outlets.  I'm beating that without a doubt. 
    
    Do you REALLY have things plugged into and turned on, in ALL of your GFI
    protected outlets? You're only "beating" the load rating of the GFI (or
    a breaker or a fuse for that matter) if you have things plugged in and
    turned on and DRAWING current. 

    I also have many outlets downline of my GFI but they are there for
    convience and are not all in use at the same time.
    
    Your problem could be a loose wire or bad connection *SOMEWHERE*
    in your house. I say *SOMEWHERE* meaning almost *ANYWHERE* in your
    house wiring due to the fact that something on one circuit affects
    something on another circuit.

    Charly
    
    P.S. What are you plugging in that is causing the problem? Maybe
    the GFI is tripping for a REASON!

.2>  Currently, it looks like the best suggestion is to split the circuit with
.2>  two separate GFI breakers, one for exterior and one for interior.  How
.2>  does that sound?

It's my understanding that breaker-box-mounted GFIs are more expensive and 
less reliable than outlet-box-mounted ones.  So perhaps you'd be happier
with a normal breaker, and a GFI outlet in the first box, for the new
circuit.  Be sure to put the GFI outlet in an place where it's easy to
check and reset it.

RE: .3

>    Do you REALLY have things plugged into and turned on, in ALL of your GFI
>    protected outlets? You're only "beating" the load rating of the GFI (or
>    a breaker or a fuse for that matter) if you have things plugged in and
>    turned on and DRAWING current.

In general, nothing on the GFI circuit was drawing when I tried the exterior
outlets.

>    Your problem could be a loose wire or bad connection *SOMEWHERE*
>    in your house. I say *SOMEWHERE* meaning almost *ANYWHERE* in your
>    house wiring due to the fact that something on one circuit affects
>    something on another circuit.

Loose wire may be one possibility but there can be others.  For instance,
a long wire run in the house is more susceptible to voltage spikes than
shorter runs, i.e. like an inductive antenna.  A good spike source can be
a motor turning on/off, etc.  The spike in turn may be large enough to trip
the GFI breaker.  There is also the possibility of ground loops.

>    P.S. What are you plugging in that is causing the problem? Maybe
>    the GFI is tripping for a REASON!

I thought the same, so I deliberately when outside with a flashlight
recharger and showed that the breaker still tripped.

RE: .4

>It's my understanding that breaker-box-mounted GFIs are more expensive and
>less reliable than outlet-box-mounted ones.  So perhaps you'd be happier
>with a normal breaker, and a GFI outlet in the first box, for the new
>circuit.  Be sure to put the GFI outlet in an place where it's easy to
>check and reset it.

Good suggestion.  I'll have to think about that.

I think its becoming clear that I've got to do some more debugging.  I
think on Saturday I'll have my wife stand in the cellar while I go plug
my electric razor all over the house!  Let you know more.

rep

    First off, sounds like the GFI breaker is bad.
    
    Reliable GFI's are not, why have a test button?  The main reason
    you hear the panel GFI breakers are not relable is due to the
    enviornment they're in.  Basements are not the nicest places to
    live.  Expense wise it costs less for the panel GFI, the other way
    you still have the expense of a breaker in the panel and two places
    to trip the circuit.  Electricians have been drilled to the point 
    of instict to place GFI's in the panel (reason being that you should 
    only have to look one place when a circuit trips).  It is much nicer
    to be pushing the GFI test button in the luxury of your own bath
    though! 
    
    As far as number of outlets per circuit...
    
    No circuit should exist with greater than an 80% load factor.
    In this case it is most likely a 15 amp circuit so 12 amps max.
    Each duplex receptacle counts for 1.5 amps so you can have 8 of
    them.  Of course, this is a general rule of thumb for calculations.
    If you have known loads on the circuit then you need to concider
    them.                                                
    
    
    Check the GFI to make sure the connections are good and clean. 
    If you still have the problem then disconnect the house wires from
    it and try connecting a load directly.  If it trips then you'll
    have your answer.
    
    Just in case you aren't sure...
    The GFI "Ground Fault Interruptor" operates by checking that current
    flows equally between both conductors (black (or red) and white).  A 
    difference of millamps of current will trip the breaker (typically 
    anywhere from 10-100 millamps depending on type).  Don't get confused 
    with grounds, phases white-wires, black-wires since they have nothing 
    to do with it in themselves.  A regular breaker checks current on only 
    one conductor (typically the black or red).
    
    I've seen cases where folks have done their own wiring and got the
    returns (whites) of two circuits crossed in the panel.  A GFI would
    not be very forgiving.  In the case I saw the two GFI circuits 
    tripped together and the guy spent hours looking straight at the
    problem.
    
    You should be able to fix this one without an OHM meter ;')
    
    -Tim 

RE: .6

Good advice.  Have you got any suggestions on why the breaker trips when
I plug something into any of my exterior plugs?  It happens on sunny days,
when the circuit is known good, and as soon as I plug it in.  The "load"
can be as little as an electric shaver.

I'm cross-posting to ELECTRO_HOBBY, I've got to believe that there truly
is something wrong, somewhere.  The circuit tripping as soon as I plug
anything in, to me, is a solid indication.

Thx. rep

    I take it you tried this and the GFI looks good?
    
    >  Check the GFI to make sure the connections are good and clean.
    >  If you still have the problem then disconnect the house wires
    >  from it and try connecting a load directly.  If it trips then 
    >  you'll have your answer.
              
    I'm not quite sure what you mean by a "known good circuit"!?!
    
    It gets messy from here because you have to start opening things
    up.
    
    Anyway, this is what it looks like you're saying.
    
    	1. The GFI is good
    	2. The inside outlets (?bathrooms?) of the circuit work just fine.
    	3. The outside ones (including garage light) trip.
        4. Other circuits appear to affect tripping.
    
    Here it comes...
    	Input to your house is single phase three wire 240V.  The goal
        for the electrician (should he realize) is to balance the load
    	as closely as possible across the transformer windings (up there
    	on the telephone pole).  The center tap allows you to have 120V
        circuits, its also the one that gets connected to earth ground.
        The goal is to have no current returning to this center tap.
    
        ie. 
    
    	 transformer
                 10A            ________ 10A   cir A
             {---------/~\------|      |-------}
             {    0A            | Fuse |-------}
             {---------/~\------|      | 10A
             {                  | Box  |-------}
    	     {---------/~\------|      |-------}
    		 10A		|______| 10A   cir B
    	            
    	A little example there.  If you tied the neutrals of cir A and
        cir B the neutral wouldn't have 20A but would be 0A, the load
        balances.
    
    	Now  the electrician might not run 120V circuits as all 2-wire.
        If there are two circuits where he could save money by running
    	one 3-wire cable instead of two 2-wire ones he'll probably run
    	the 3-wire (I'm not counting grounds here).
    
        If your GFI circuit was done this way then it would explain
        the weirdness.  Then again, maybe the neutrals got mixed up
        out there somewhere.  You'd have to take every box apart
        and trace out the wiring, messy.
                                                
    This is a pretty far fetched problem, but possible nontheless.
    
    So, what have you actually tried?  What does the circuit really
    look like and how old.

RE: .8

>    I take it you tried this and the GFI looks good?

I haven't done that yet.  I wasn't clear whether or not you were saying
that this may be a reason as to why the breaker trips when I plug something
in.  I'll take a look at it.

>    I'm not quite sure what you mean by a "known good circuit"!?!

"Known good" means that I can plug things into my bathroom outlets, or turn
on my exterior lights and the circuit will not trip.  Its only when I use
an exterior plug that the breaker trips, or in the instance of some wierd
non-deterministic failure, i.e. the bathroom plug was fine before I went
to sleep but when I woke up, it didn't want to work anymore.

>    So, what have you actually tried?  What does the circuit really
>    look like and how old.

The wiring for one of the bathrooms, is less than 1 year old.  The house
is roughly 10 years old.  I bought the home last year, just as the new
bathroom was being completed.  I'd venture to say, that the previous owner
introduced the problem in the wiring of the new bathroom.

I haven't tried much, because I was hoping for a definite answer given the
symptoms I've described.  It looks like its going to take some more digging,
and I'll be back with more data in the near future.

Thx. for all your suggestions,
rep

    FYI:  Those GFI breakers don't appear to be cheap.  A friend bought
    one this past weekend at Marlboro Electric and had to shell out
    ~$42 for one.  

    GFI breakers are subject to "nuisance" tripping if the wire run
    from the breaker is to long.  I believe this is in the range of
    75 to 100 feet. You might want to check this out.  Also, you want
    to be sure all joints and connections in the circuit are secure,
    electrical and mechanical.                                    

.11>    GFI breakers are subject to "nuisance" tripping if the wire run
.11>    from the breaker is to long.
    
    	Why would this trip a GFI breaker?
    
    -mike z

    I am thinking about connecting a GFI outlet to two existing switches
    in my bathroom. I would like to keep all three together in a triple
    box. Does anyone know if anyone sells a three box switchplate with
    two switches and a GFI outlet?
    
    Mike
    

    There are a couple of different possibilities.  One is to replace
    the switches with those like the Leviton "Decora" line that have
    large rectangular openings like many GFI outlets, then use a triple
    plate designed for those.  You can also get GFI outlets that fit
    a normal two-hole outlet cover, and it should be simple to find
    a cover for that.  Go to your local electrical supply store (a
    REAL one, not just a hardware store), and you will find a large
    selection of outlet/switch covers.
    
    				Steve

    Try an electrical supply house.   There's one in Fitchburg that has
    them for sure (but I can't remember the name).   Wheelen Supply in
    Gardner would even special order some special GFI-related covers. 

      If  your home has circuit breakers (not fuses) you can replace the
      circuit breaker for the circuit the bathroom is on with a GFI type
      breaker.   This  might  be  an easier retro-fit for about the same
      money. Our bathrooms, kitchen and outside outlets (three different
      circuits) were originally wired this way.

    Read notes 574.5, 816.16, 1659.13 before considering this approach.
    
    Charly

    re. .12
    
    You're not dealing with perfect conductors and as a result there is
    an imbalance (cable resistance) between the hot and neutral legs of
    the circuit.  As the length of the wire run increases it is possible 
    for this imbalance to be greater than the trip point of the GFI
    breaker.  In the area of nuisance trips it is related to circuit load, 
    outside temperature, wire routing (tight bends...), etc.
    
    This is my understanding of the problem.  Someone else might have
    a better or more technically accurate explanation.
    
    /Ernie

      There are two non-issues in the notes referenced by the preceeding
      reply:  (1) cost -- Breaker-box vs.  coubined-outlet GFIs are  not
      different  enough  in cost for this to be a serious consideration.
      (2) Reliability is a function of the conditions, not the  type  of
      breaker.  If _ANY_ GFI trips frequently YOU HAVE A PROBLEM. Either
      the GFI or installation/wiring is faulty, or you are  usinging  it
      in an unsafe way.  
      
      Go with whichever is easier to install.

    I need to put an outlet in the bathroom anyway because we don't
    have one in there now so I might as well just install the GFI outlet
    in the bathroom.
    
    Thanks for all the help!
    
    Mike
    

	This is probably mentioned in one of the other GFCI notes,
	I didn't look throught them all.  I believe that it is
	recommended that you trip the GFCI every month, this is easier
	to do for you or others less interested in things electrical
	if its easily available... also you are reminded every time
	you use that outlet.

    .re .7 (tried to minimize quoting....!) 
    
    > I finished a basement playroom for my kids - put in ONLY GFCI-protected 
    > outlets since GFCI's provide some protection from electrocution - 
    > (but they aren't 'perfect' since you can electrocute yourself across 
    > the hot and neutral without creating a ground fault). I figured the 
    > most likely case is metal-object-poked-into-hot-side-of-outlet which 
    > will trip a GFCI.

    I put in a GFCI Circuit Breaker in my basement for a new laundry
    room.  How can I know it's working?  Does the test button actually
    create a ground fault, and if so, will it not trip the circuit if
    it is installed incorrectly?  Is there any way to test it?
    
    The breaker works fine, but I DID get a little jolt when I was tightening
    a screw on a live outlet (yes, stupid, wasn't it?).  The breaker
    didn't trip.  I hope it would have if it was a serious, life
    endangering shock.  
    
    Any ideas, comments?

    
    
    	Good question! 
    
    	I was wondering the same thing myself. I have a situation where
    I can't install a GFCI outlet due to overcrowding the box it has
    to go into so I'll end up having to go with a pannel breaker. But
    I always wondered how to really make sure outlets down stream from
    the GFCI outlet or breaker are indeed protected. I did manage to
    trip on once do to plugging in a extention cord that had gotten
    wet but thats the only time I ever had a REAL trip!
    
    			...Dave
    

    RE: Testing GFI outlets and breakers.
    
    	I had this same question when I replaced our old fuse service
    	entry with a 30+ circuit breaker panel.  To bring it up to code,
    	I had to (among other things) install about 10 GFI breakers.
    	(Not knowing how the house was wired, I didn't install the GFI
    	outlets.)
    
    	A real easy to test them is to short the neutral line to ground.
    	When you do this the breaker/outlet should trip.  If it doesn't,
    	you probably have problems.  Note that grounding the hot line
    	will also trip the breaker :-), but won't test the GFI capability.
    
    - Mark

    
    
    	Just so we don't end up with a bunch of dead/fried DECies, What's
    a safe way to 'short out' an outlet so as not to cause sparks, flames
    and injured people? 
    	
    			...Dave
    

I bought a plug-in electrical tester for about 10$ which also has a 
little button on it which when pressed causes a slight short which causes
the GFI to trip.  You can try it on any outlet and it will trip the GFI
if it is upstream. 

I would replace the box with a larger one so that you can put in a GFI 
outlet since they are MUCH cheaper than GFI circuit breakers and much more
reliable.

gary

    re .49:
    
    I would assume you need to short neutral to ground, then apply a
    load downline, in order to trip the GFCI. According to my
    understanding, shorting neutral to ground on an open circuit should not
    trip the GFCI, because no current flows. Is that correct?

    A better test of the sensitivity of a GFCI would be to connect an
    appropriate resistor from hot to ground, allowing you to control
    the leakage current (watch the power rating on the resistor).
        
    For safety, you should make your connections inside a screw-togehter
    3-prong plug assemply, then button it up and plug into an outlet.
    

    RE: .52
    
    	Hmmm.  The breakers I have (Gould/ITE) would trip when I shorted
    	neutral to ground, without a load.  Sounds like it is about
    	time I check out some reference material to find out how those
    	little things work.
    
    - Mark

    The GFCI discussion raises a question: exactly what happens when
    you push the "TEST" button on a GFCI outlet?
    
    pbm
    

    
    
    	Getting back to the original question does any one know how
    these thinks realy work? and are willing to type a lenthy explanation?
    Get as technical as you want. 
    
    	Re back a few:
    
    		What size load do you need? I take it that you don't
    have to pull 15 amps to get the GFI to trip? What is the min current
    that should trip these?
    
    				...Dave
    

    Shorting neutral to ground does trip the GFCI. Do this by simply
    shorting the taller slot (next pin on receptacle in a clockwise
    rotation from the ground slot) to the ground slot. 
    
    After accidentally discovering this, I decided to try the same with
    the hot side while wiring inside a metal box; DON'T DO IT!  If you
    are watching what you are doing, you will see a spot everywhere
    you look for the next hour or so... from the intensity of the arc
    that you will draw by doing this!  Yes, this did trip the GFCI,
    and also its supply breaker.
    
    Good idea about totally enclosing your "test fixture", but play
    it safe; dont' try to bleed current through the hot side, the neutral
    is alot safer to muck around with.
    
    BTW--I had a GFCI that would not trip anymore this past fall, so
    make sure you do test yours, to ensure that it still does work.
    
    Mike

    No, you don't need 15 amps to get the GFI to trip.  You better not! 
    The minimum current that should trip a GFI is somewhere around 50
    milliamps, since that's the amount of current that, properly applied,
    can kill you!  Remember, the GFI is designed to detect and trip if a
    *leakage* current is detected.
    
    The Leviton(sp?) GFI I have is speced at 50 milliamps leakage current.
    
    Jim

        Re "Why it works shorting neutral to ground."
                
                
                In a system where there is nothing drawing any current
        from your supply panel (where the ground and neutral are tied
        together), it probably wouldn't work. But since that condition
        is almost never is true, there will be enough voltage difference
        between ground and neutral from the normal IR drop in the
        neutral wire to cause enough current to flow in the ground wire
        to trip the GFI.
                
                /s/     Bob

    I'm sorry if this seems like a digression, but WHY is there usually
    some voltage potential between neutral and ground?  After all, they're
    both connected to the same block at the breaker panel, aren't they?
    
    			Steve

                If you have any current in the neutral (which you will
        with a load applied), then you have IR drop in the neutral. You
        do not have this voltage created in the ground as there is no
        current flow in the ground. 
                
                /s/     Bob

>    I put in a GFCI Circuit Breaker in my basement for a new laundry
>    room.  How can I know it's working?  Does the test button actually
>    create a ground fault, and if so, will it not trip the circuit if
>    it is installed incorrectly?  Is there any way to test it?


I recently bought a house which has a GFCI receptical in the bathroom.  The
home inspector I used (Cornell) pointed out that even though pushing the TEST
button caused it to trip, the power was not being disengaged.  (I never would
have thought to test this.)  Anyway, I bought a replacement but when I went to
install it I found that the original had been installed incorrectly.  Power was
connected to the LOAD wires instead of the LINE wires.  I just reinstalled the
existing outlet in the correct manner and it works correctly now.

In summary:

So even though the test button may trip the breaker, this doesn't ensure that
power has been disengaged.  This situation can be caused by incorrect
installation.

-Kurt*

re .61

This is probably the most common mistake made with GFI circuits. On the 
surface it usually seems pretty straightforward, however, you'd 
probably be surprised how easy it is to screw up an installation. 

Not that I've ever made any mistakes though. ;-)

chuck

		As someone  suggested in a earlier note a better solution 
	to my over crowded box and GFI breaker  problem would be a bigger 
	box as opposed to a GFI breaker in the circuit panel.  My biggest 
	problem is that I don't really have the room for  a  bigger  box.  
	Unless  of  course  someone  knows  of  a  box  that will fit  my 
	application as sown below?  
	
		|*|*|      |*|*|*|
	        |*|*|      |*|*|*|
	        |*|*|+----+|*|*|*|
	        |*|*||    ||*|*|*|
	        |*|*||    ||*|*|*|
	        |*|*||    ||*|*|*|
	        |*|*||    ||*|*|*|
		|*|*|+----+|*|*|*|
		  ^    ^      ^
		  |    |      |
	     (2)2x4s  box   (3)2x4s

		Now I can't re-wire so as to put the GFI in a  area where 
	there is room for a bigger box.  The room is almost  all  windows 
	and the framing  leaves  almost no empty wall cavities.  Besides, 
	this box location is  the most convenient for the GFI (right next 
	to door going out to  deck).    So has anyone seen a box which is 
	the standard 2 or 2&� inch width but taller and come with a plate 
	to reduce the opening for installation of a outlet?
		
		As one can  see  the  SIMPLEST  solution  is to use a GFI 
	breaker.  This was my reason for wanting to know  how  to  do  an 
	external test of the GFI breaker and with all the bad press about 
	these  types  of  breakers  I  want  to  make sure the circuit is 
	working properly.		

	       	Now if I understand all the comments  made here correctly 
	the best way to test the GFI would be  to  build a little plug in 
	adapter wired as follows:
	
		HOT        NEUTRAL 		  
		|
		|
		|
		+---\/\/\/\/--- GROUND	
		  2.4K ohms =50ma
		  10 watts
			
		This would  most  likely  imitate the undesired condition 
	which  would cause electrocution, with a person being replaced by 
	the resister.  Even though most said to connect neutral to ground 
	this  seems to  me  to  be  of  very  little  danger  in  a  real 
	environment as there should  under  most  circumstances  be  very 
	little potential between neutral and ground.  If I really want to 
	make  sure the GFI works the  real  test  would  be  to  run  the 
	resister from HOT to a nearby water  pipe  rather than the outlet 
	ground.  Any comments?
	
		...Dave

    Dave,
    
    Why don't you install a GFI outlet someplace else (bath, garage,
    workshop, etc...) and then wire this outlet downline from the GFI
    protected outlet? 
    
    Charly

    I don't know if this is up to code, but, a WIREMOLD (tm) surface-mount
    single-gang box would extend the box out enough to give extra cubic
    space inside for all the wiring. I'm doing this exact thing, and
    have the box already, just haven't put it together yet.
    
    The GFCI receptacle mounts in the WIREMOLD (tm) box, so it sticks
    out some. In my case, it's not a problem; it's in the garage...
    
    Mike

    
    I've seen some metal boxes that appear to be almost 3" deep.  Will
    that be big enough?  The easiest way was stated in .64 though.
    
    

    RE .64 AND .66
    
    What do you mean install a GFCI downline?  Do you mean install one
    between the breaker and the spot that is being considered?  What
    will that do?  When a person grounds the one outlet, will the GFCI
    that is downline (actually upline?) trip and shut off the electricity
    to the whole line?  Isn't this what a GFCI breaker is, theoretically,
    supposed to do?
    
    Ed..
    

                A GFI outlet usually has four terminals (or wires) on
        it. Two are labeled LINE and are to be connected to the line
        coming from the source (panel). The other two are labeled LOAD
        and are to be connected to any of the circuit located beyond the
        GFI outlet. In most (if not all) GFI outlets this means that the
        items (outlets usually) located beyond the GFI (away from the
        panel) are also protected by the GFI. 
                
                Yes, this is the same as a GFI breaker does. However, it
        seems that outlet style GFI's are more robust in manufacture and
        tend to last longer than GFI breakers (why this is so, I don't
        know - I haven't heard a coherent reason).
                
                /s/     Bob

	RE: < Note 1659.67 by AKOV13::FULTZ "ED FULTZ" >

	GFI recep. have 5 terminals (or leads in some cases on them). One is
	for ground.  Two are marked LINE.  This is where the HOT and neutral
	wires go that are coming from the panel.  The other two are marked 
	LOAD.  You can run a line from these terminals to another recep.
	"down line" and that recep. will be GFI'ed also.  If you get these
	reversed the down line recep will work but will not trip the 
	GFI when tested.   This is very useful in large bathrooms that
	require more than one recep.  You bring your LINE into one and then
	feed your other recep from the LOAD side of the GFI.  If it trips,
	all you have to do is reset it at the recep (you don't have to go
	down to the circuit panel).  There are several other applications
	that make GFI recep more desireable than GFI breakers.

	Yes this does make the GFI recep. act like a GFI breaker.

	bjm

    One of the best reasons to use a GFI receptacle - COST. You can
    always find one on sale for $9-$10. A GFI breaker usually runs 
    close to $40. Also, it is nice not to have to walk down to the
    cellar if one trips.
             						Chris

    
    I want to install a GFI in a bathroom.  I will run a new circuit
    to the bathroom.  Should it be a 15 or 20 amp line?  The GFI is
    rated for 15 amp but a 20 amp feed thru.  Why 2 different ratings?
    
    Phil

    The outlet itself in the GFI is 15 amps (if the outlet looks like
    this:  | | ,its 15 amp.
           | |
           | |             
    
    If it looks like this: | |  its a 20 amp.
                           | |-
                           | |               
    
                               
    The feedthrough is the current that can pass through the GFI, not
    the capacity of the GFI's outlet itself.
    
    Eric 

    
    OK, but what gauge wire should I use to wire it?  12 or 14?  Should
    I wire it for the outlet or the pass thru?
    
    Phil

>< Note 1659.73 by VINO::GRANSEWICZ "Which way to Tahiti?" >
>                    -< But the $64,000 question remains... >-
>
>    
>    OK, but what gauge wire should I use to wire it?  12 or 14?  Should
>    I wire it for the outlet or the pass thru?
>    
>    Phil



Wire in a 15 amp branch circuit using 14 gauge romex.

Ross

    what good is it to have a 20 amp pass thru if you wire it 
    for 15 amps.. or did i miss something ??
    
    tony

    I asked the same question and was told it was for surge protection.
     The wire going into and out of a 15amp rated GFI is 14 gauge.
    You must use the same gauge wire,14 gauge,to connect it into a
    circut.
     Acording to my local Wire Inspector,the wirerunmust be of the same
    gauge wire.
    		Wayne

    
    RE: .75
    
    Exactly my question!  I checked the outlets this morning and they
    have stickers on the side which say "Rated 20 amps".  Less conspicuous
    is the stamp on the metal surrounding the outlet which says 15 amp.
    Now how can this thing have 2 different ratings???
    
    Also, the instructions clearly state that the GFI is NOT AN OVERCURRENT
    DEVICE.  So you would need a 15 amp breaker to protect the outlet
    I suppose.  And this means 14 gauge wire.  Seems pretty logical
    but why do they through in the "20 amp feed through" if it's
    meaningless?  And then there is the sentence which says to connect
    the outlet to Aluminum wire using 12 gauge.
    
    Can anybody figure this out or should I call the company?  What
    does the code say about these things?
    
    Phil

    RE: .76
    
>   I asked the same question and was told it was for surge protection.
    
    What is "it"?  I don't see where surge protection comes into this.
    Could you explain a little bit more?
    
    Phil

                       -< Add more confusion to confusion... >-

    
	Since when does the code specify a MAXIMUM wire size?  For
	15 amp circuits, you have to use AT LEAST 14 gauge.  Some
	recommend using 12 gauge even though it's not required.
	It doesn't cost much more (except 12-3), gives a lower
	voltage drop when the load is high, and if you ever want to
	change it to 20 amps, you don't have to rewire.  (You just
	have to replace all the 15 amp-rated outlets and such...)

    I don't know if I can succeed in clearing up the confusion here, but
    I'll try.
    
    Forget GFIs for the moment, and note that it's common practice (and
    perfectly legal) to connect multiple 15A outlets to a 20A circuit, i.e.
    to a circuit with a 20A breaker and #12 wire.  Your kitchen is probably
    wired this way, in fact.  The idea is that the circuit can support
    multiple devices totalling up to 20A, but the devices on any one outlet
    won't be more than 15A.  So, for example, you can run an 8A toaster and
    a 10A frying pan on different outlets of the same circuit, but running
    them on the same outlet would exceed that outlet's capacity.
    
    These multiple outlets are wired in a daisy-chained manner. Electricity 
    travels from the breaker panel to the first box in line.  There it
    splits:  some goes to the outlet there, and the rest goes to the other
    boxes (and outlets) on the circuit.  While the outlet in the first box
    only needs to handle 15A, the connections there need to handle the full
    20A.  That's why using the outlet's screws and the little metal strip
    between them to make pass-through connections is a bad idea, and why
    using the push-in connections is even worse.  The preferred technique
    is to wire-nut three wires together - the supply, the wire to the next
    box, and a tap to the outlet - so the full 20A never gets near the
    outlet itself.
    
    Now back to GFIs.  Assume that you want to use one GFI to protect
    all of the outlets in the circuit described above, so you'll install
    the GFI in the first box in line.  In order to bestow GFI protection on
    the downstream outlets, you need to wire things in such a way that the
    full circuit load flows through the GFI package - the split described
    above still happens, but it happens _inside_ the GFI package.  Although
    only 15A can be plugged directly into the GFI outlet, supporting the
    downstream outlets requires that the GFI's internal wiring be able to
    handle the full 20A.
    
    The "15A direct/20A flow-through" rating simply indicates that the
    GFI you bought is capable of supporting the above circuit configuration.  
    Since you didn't mention any downstream outlets, I suspect that none of
    this discussion applies to your installation.  If you'll be wiring the
    GFI to a dedicated circuit, a 15A circuit (15A breaker, #14 wire) is
    sufficient, although a 20A circuit would allow more flexibility for
    future changes.

Geez, Some of you guys like making a mountain out of a mole hill!
How about this explanation: Just about any switch, receptacle, etc. for
the home is rated for 20 amps. They can be used in 15 or 20 amp branch circuits.
As stated earlier in this note, receptacles can be purchased to accept either
15 or 20 amp plugs. The GFI receptacle is rated at 20 amps, but accepts only 15
amp configured plugs:





		|	|			|	|
		|	|			|	---
                                                        |
		    O		15 amp		    O            20 amp


Excuse my poor drawings, but you should get the idea.
In a bathroom circuit, there is no reason to run a 20 amp circuit, unless you
like wasting money on 12 guage wire.

Ross

    
     rep .80 
    
     Thanks,I could not have explained any better!
    
    rep.81. 
     Not true! It depends on what the bathroom has in it for lighting,over-
    head deater/infared bulbs and such. I am putting a overhead light with
    a heater,fan,main light and night light in our bathroom. Now if I 
    attach this to a 15a circuit and my wife runs the hair drier with
    the fan and heater going- CLICK- no juice! I am not busting your chops
    but this is my real life situation.
    
     The "Flow through" of 20a in a 15a GFI is best explained by .80. I
    called the inspector back and he verified it.I must have heard him
    incorrectly.... 
    
     As far as different size wire in a single circuit goes,The wire must
    step down in size (12 to 14 gauge) and not the other direction 
    (14 to 12 gauge).  Make sense?
    
    		Wayne

    
>    rep.81. 
>     Not true! It depends on what the bathroom has in it for lighting,over-
>    head deater/infared bulbs and such. I am putting a overhead light with
>    a heater,fan,main light and night light in our bathroom. Now if I 
>    attach this to a 15a circuit and my wife runs the hair drier with
>    the fan and heater going- CLICK- no juice! I am not busting your chops
>    but this is my real life situation.
    
     Calculate your load and figure out what you need. An ordinary (configured)
bathroom can easily get by with a 15 amp circuit.

Ross

    
    
    	If you're really conserned about it spend a few extra bucks,
    	go down to the local electric supply store and box a GFI outlet
    	with the 20 amp configured plug and rating. They do have them,
    	all the outlets buy sinks in MRO4 are setup with them. I've
    	been meaning to get one for the dedicated 20 amp circuit I put
    	in my garage to run power tools.
    
    					...Dave
    
    

How about this scenario: A person cooking in the kitchen that has the minimum
(2) 20 amp small appliance circuits with 15 amp "configure" duplex receptacles.
The person cooking plugs in an electric fry pan that at high temp setting
draws say 12.5 amps. This same person plugs in a toaster on the same duplex 
that draws maybe 6 amps. The breaker is within it's threshold of approx. 20
amps and doesn't trip. Does the 15 amp receptacle do a melt down? Not likely
Is this a legal branch circuit? Yes

			Table 210-21(b)(2)
	Maximum Cord_ and Plug-Connected Load to Receptacle

________________________________________________________________________________
________________________________________________________________________________

Circuit Rating			Receptacle Rating		Maximum Load
Amperes				Amperes				Amperes

15 or 20			15				12



This (I believe) is a grey area in the code. The described scenario is a code
legal branch circuit. Does the average homeowner know if he's pulling more
than 12 amps out of a receptacle? I doubt it.
(My thought) The manufacturers build in the extra capacity in the devices.
I'm open to argument and thoughts on this one.
BTW, I don't agree with everything in RE:80. When time permits, I'll get into 
it.


Ross

     I have a melted down 15a gfi duplex outlet. It came off of temp.
    power pole. I am building a house as of now. It is attached to a 
    20a breaker. The poor sucker died at night while i was working
    in the house. 
     Bear in ming that this is my only source of AC and I was abusing
    the heck out of it!! The 20a breaker never tripped but the 15a
    GFI with 20a carry through melted down!!  
    
     I replaced it with 20a GFI.
    
    		Wayne

    
    Realize that you may need a larger than standard box to house a
    20A GFI outlet and it's wiring. They are physically larger than
    the 15A versions, at least the Leviton brand.
    
    CdH
    

>< Note 1659.86 by VLNVAX::HEDERSTEDT "T.B.S." >
>                         -< I will send it to you??! >-
>
>     I have a melted down 15a gfi duplex outlet. It came off of temp.
    

In my scenario, I had standard duplex receptacles in mind. However, your
GFCI outlet might of been defective or you were drawing more than 20 amps?
BTW, I'm not a fan of GFCI outlets. I rarely use them due to the fact they're
more trouble prone than GFCI breakers. They're widely used now because they're
less expensive.


Ross

     I should not have been drawing more that 20amps or i should have
    tripped the 20amp breaker. I have 200'+ of ext. cords to the house.
     They are 12awg but the draw + voltage drop killed the GFI.
    
     In my house,since its being built and I am wiring it,the GFI's are
    in the main panel.I do not care for the GFI's that are in the outlet
    boxes. Any GFI breakers that are in my house are 20a.I would rather
    be safe than sorry with the extra $'s invested in 20a service to
    certain places in my house. The bathroom is one and the kitchen
    is the other. By code I do not have too much of a choice as far as
    the GFI's go but I plan to add a bit of fudge factor (extra amps)
    to these area's. I know Mr. Murphy and his law's will bite me
    in the backside later and I will be pissed that I did not do it
    right (for my usage) the first time.
    
     Six or 10 bucks now is cheap compared to running another line at
    another date!
    
    		Wayne 

>< Note 1659.89 by VLNVAX::HEDERSTEDT "T.B.S." >
                               
No one say's you can't oversize your circuit. That's up to you.
BTW, don't forget to ad GFCI for the garage outlets, (1) basement outlet, and
outdoor outlets, also.

Ross

    
      No problem there, already done.Except the garage..Can't afford it
    yet.    Maybe if I did not buy 12-2 and bought 14-2......:^)
       enough said!
    
    		Wayne

    re .88, last few
    
    I was led to believe (I thought I read it here) that the GFI outlets
    were more reliable than the GFI circuit breakers.
    
    Which is which ??

    re .92:
    
    Experience dictates that the outlets are more reliable than the
    breakers.
    
    A couple of years ago we had a lightning strike between my house
    and my neighbor's. Outlet GFCIs in both houses came through fine.
    A GFCI breaker in his house bit the dust. An elecrtician buddy of
    his advised using an outlet replacement if possible - because they're
    more reliable.
   

    The 20a GFCIs do not need a larger box than the 15a.  I have 15a in the
    Bath, but there are 2 20a circuits into the garage/shop, and those must
    also be GFCI, and so I had to get 20a GFCIs.  No problem with size. 
    The COST, however, is much more - I seem to remember it being around
    $20 or so each, where the 15a typically run about $9.99 at most places.
    I did find that one color was more expensive than the other (brown or
    white, I cannot remember which was higher), which I thought odd.
    
    I agree with .-? on the 15 vs 20a for the bath - 15a is sufficient
    unless you have something like a baseboard (non 22v) heater in there.
    However, what you will find in a lot of houses is that the bath is not
    on a seperate circuit.  I ended up rewiring the bath to have its own
    because the power would trip when I dried my hair in the summertime
    when the attic fan was running at the same time and the lites happened
    to be on in two adjacent rooms!
    
    

    
>    Experience dictates that the outlets are more reliable than the
>    breakers.
    
    My experience has been the opposite. Breakers can take a hell of a lot
more abuse and have a longer longevity. Contractors like the GFCI outlets 
because they're $25+ dollars cheaper than using the breakers in a circuit!

Ross  

    DISCLAIMER:
    This is all my interpretation and may bear no relationship to reality.
    
    I don't think the cost has anything to do with the reliability/quality
    of the breaker vs. outlet.  From the description of the GFI, it
    performs a different function than a regular breaker.  Combine the
    GFI capability with a breaker and you now have 1 device performing
    2 functions thus more cost.  Also, breakers can "stick" (even though
    everybody toggles theirs once a month ;-)) and you'd be left without
    any protection if this happens on a GFI breaker, right?  But I guess
    any device coul;d fail for any number of reasons.  I would think
    that the more complicated the device (more functions), the more
    they are prone to fail.
    
    End of GFI cost/quality theory...

    

    
    20A GFCI outlets are larger than 15A in certain brands. If you intend
    to hook up a parallel set of outlets downstream then when using
    12G wire they may not fit into a standard box and still *meet Code*.
    I had to use an extension box on my installation.
    
    CdH
    

    I meant to mention that there are brands that do not use pigtails, but
    have screws on them, which reduces the amount of space in the back of
    the box needed to make connections.

    
     I did some more checking on this and the 20a service to the bathroom..
       I know,what a nit-picker. I was with the Electrical Inspector for
    several hours yesterday. I Was enlightened quite a bit by his
    knowledge. No armchair coach here!
     Any circuits should not be loaded more than 80%. If someone
    turns on a 2000 watt hair drier,that happens to be about 16.6amps.
    Now if the lights are on the same circuit and are on as well......
    
       The Inspector mentioned that it won't be long before code requires
    20a service to a bath. 
    
     I replaced a 15a gfi with a 20a gfi. It will fit in the box but it
    did take up more room! In act,the gfi will not quite go in all the way
    because of the ridgid pipe clamp that protrudes into the box.
    
         wayne

>    Any circuits should not be loaded more than 80%. If someone
>    turns on a 2000 watt hair drier,that happens to be about 16.6amps.
>    Now if the lights are on the same circuit and are on as well......
    
      2000 watt hair dryer? Does it heat the room as well as burn hair off
your head. \\8*} There's a chart in the code that gives max load per outlet.
See a couple of replies back. I'm not too worried with my little 15 amp 
bathroom circuit, my Wife has every electric hair gadget known to man and
I've had no problems.
BTW, Ask the inspector the question I asked a few back: How does the average
homeowner know if he's exceeding 80%?

Ross

      The 2kwatt is NOT uncommon.. Most go up to 1500 on hi setting.Circiut
    breakers trip at @120% load,not at 100%.  As far as the 80% load goes,
    Most people cannot tell!  Where that comes in is NEW construction and
    good planning.  I may not need the full 20a in my down stairs bath but
    I will need it for the upstairs bath/master bath!  I plan for worst
    case involving what might be running in the bath at one time not
    what I can get by on.  Poor planning will get back to at the worst
    time possible.   I am not over-building the circuit in the bath,just
    adding up wattages used for each appliance that my be on at the same
    time and that figure tells me 20a.
    
       Wayne

>< Note 1659.101 by MRVAX::HEDERSTEDT "T.B.S." >


 >     The 2kwatt is NOT uncommon.. Most go up to 1500 on hi setting.Circiut
    
Take a look at the manufacturer's full load rating on the appliance. Not
the phoney baloney decal number. If you have something that draws 16.6 amps,
it's a heat gun, not a hair dryer.
Btw, Let's take this argument offline.

Ross

    My big mouth,here it go`s.
    You all wire your house the way you want but,I use 12/2 wire less
    on the braker then it calls for so the wire in the walls will not
    git hot,the braker will go before a fire comes on.
    I use ohms law to see what I can plug into the lines.
    I use 1 braker for each room and 1 each for frigh,stove and etc.
    If you use 15a brakers do the box calls for 20a and 12/2 wire you
    will not need a GFI.
    The wire and box can`t get hot from overload.
    
    Even if you don`t use ohms law you can`t plug more in because the
    braker will go first.
    If you do this again and again you need help.
    
    Go ahead tell me I`m wrong.....I`m all your`s,Give me h###.
    

>    If you use 15a brakers do the box calls for 20a and 12/2 wire you
>    will not need a GFI.
>    The wire and box can`t get hot from overload.                  
    
    I'm sorry.  I don't understand what this means.  Could you restate
    it please.

    .103:
    
�    If you use 15a brakers do the box calls for 20a and 12/2 wire you
�    will not need a GFI.
    
    If I can say this without sounding nasty...
    
    GFCIs and "over-engineering" are two different things.  GFCIs are meant
    to trip when they detect more than a certain amount of current (50
    milliamps?) going to ground;  the intent is to protect humans from
    lethal shocks, usually in situations where there is water around to
    help make a nice conductive path from the hot side of an outlet to
    ground.  (Note that they are required for outlets in bathrooms, in
    kitchens, and exterior outlets -- all places where you're likely to see
    water splashing about.)
    
    
    IMO, "Over-engineering" or "over-building" isn't such a bad thing, although
    it shouldn't be necessary if you're following the code.  As long as the
    inspectors don't beat you up for putting a 15A breaker on what would
    otherwise be good for 20A, *and nobody tries to draw 18A because the
    outlet says 20A on it*, you should be OK;  it's more expensive, but
    the extra expense is buying some peace of mind.
    
    Dick

    I usually skip the electrical notes, but isn't it the grounded pipes,
    not the splashing water, that makes bathrooms and kitchens dangerous,
    electrically speaking?

    
    RE: .106
    
    No.  It's the electricity...
    

    .106:
    
    Yeah, now that you mention it, it's probably the pipes more than water;
    although if I had my hand in the sink and got zapped, I probably
    wouldn't care much whether I was clutching a faucet, or just had my
    hand submerged...
    
    Dick

		As I  recollect from my high school chemistry class water 
	in it's pure form (H20) is a  very  poor conductor of electricity 
	it's the minerals in water that make it  conduct.   As I remember 
	the demonstration the professor stuck his hand in a  tank of very 
	pure (H20) with electric probes inserted in the water, and  lived 
	to tell about it.
	

    				...Dave
    	

    Dragging up this old topic since I just started reading this
    conference... 
    
    Actually, a resistive imbalance between the hot and neutral legs
    wouldn't cause the GFI to trip.  As long as there isn't any current
    leaking out of the circuit, the GFI stays untripped.  In the real
    world, there is always some current leakage associated with wiring.
    The more wiring present (due to longer runs, etc.), the more leakage
    exists.  This leakage biases the GFI toward its trip point, increasing
    the likelihood of nuisance tripping.
    
    Don't be shy about returning a GFI which exhibits a nuisance tripping
    problem.  You won't be the first one to do so by any means.  In
    the early history of GFI production, return rates for nuisance tripping
    were often in the 25-50% range!  Nuisance tripping continues to
    be the single biggest problem for GFI manufacturers.
    
    GFIs must trip with faults of 5ma, but the trip time for such
    a small fault is permitted to be in the range of several seconds.
    As the fault current increases, the required trip time decreases.
    With faults in the 100ma and above range, GFIs trip in well under
    100ms.
    
    The test button is there by mandate of UL, not because GFIs are
    inherently unreliable.  This device is supposed to be able to be 
    trusted to save your life, so that ability should be verifiable.
    The fault injected by the test current is usually between 5 and
    10ma, thus it takes a few seconds for most GFIs to respond to a
    press of the test button.
    
    								Marc
                                                                

Re: .80 -

>     So, for example, you can run an 8A toaster and
>     a 10A frying pan on different outlets of the same circuit, but running
>     them on the same outlet would exceed that outlet's capacity.

    So where's the protection against this situation?  I don't understand
    what happens when you exceed the outlet's 15A capacity when the outlet
    is on a 20A circuit.  

Re .81 -

> How about this explanation: Just about any switch, receptacle, etc. for
> the home is rated for 20 amps. They can be used in 15 or 20 amp branch 
> circuits.
 
    Then why are some marked "rated capacity 15A"?  If the only 
    difference is the plug configuration  (and I don't think that's 
    true. The 20A outlets seem "heavier"), why are they differentiated
    in the first place? 

    Still confused, especially since the 15A outlet/20A feed-through
    GFCI do in fact, exist.

    Barry

>< Note 1659.110 by LEVERS::COLELLA "Now enjoying Area Code 508" >
                 -< When the outlet melts, you're at 15.1A... >-

>    So where's the protection against this situation?  I don't understand
>    what happens when you exceed the outlet's 15A capacity when the outlet
>    is on a 20A circuit.  
>
The code allows (rated) 15amp outlets on a 20 amp branch circuit. The idea is
that the Homeowner will/(should) not exceed 12 amps on one device.
Can 12 amps be exceeded? Sure. Will the 15 amp device do a meltdown? Not 
likely.


>Re .81 -
>    Then why are some marked "rated capacity 15A"?  If the only 
>    difference is the plug configuration  (and I don't think that's 
>    true. The 20A outlets seem "heavier"), why are they differentiated
>    in the first place? 
>
Yes, The 20a device is more substantial, but 15 amp devices are adequate in 
most home branch circuits. Save the 20a outlet for the Mig welder circuit in 
the garage.

>    Still confused, especially since the 15A outlet/20A feed-through
>    GFCI do in fact, exist.
>
>    Barry

    < Note 1659.111 by MAMIE::THOMS "Ross - 264-6457" >
                                  -< Read on >-

>The code allows (rated) 15amp outlets on a 20 amp branch circuit. The idea is
>that the Homeowner will/(should) not exceed 12 amps on one device.
>Can 12 amps be exceeded? Sure. Will the 15 amp device do a meltdown? Not 
>likely.

    This seems very strange.  The code is now relying on the good sense
    of the user? (!)  I agree the outlet wont meltdown, but if it can really
    handle 20A, why rate it at 15?  I guess I'm beating a dead horse
    here, but it still seems to me that everything on the other side
    of a 20A fuse should be able to handle hours and hours of 20A
    current flow without getting very hot.  If 15A receptacles can handle
    that, then call them "20A", if not, they (In My Humble Opinion) 
    shouldn't be on a 20A circuit.  Apparently, the Code and the makers
    of GFCI 20A flow/15A outlets  don't agree.

    Barry    

    Press next unseen to avoid lengthy technical discussion of GFI
operation.

    Being new to this conference, I just found this note.  Since there
appear to remain questions about GFI theory of operation, I'll present
what I know for public consumption.

    Behold my crude combination block/schematic diagram of a typical GFI:


                                               |  Push to test
                                      ~16K   -----
                      +--------------/\/\/\--O   O-+
                      |     ___          ___       |
                      |    /   \Diff    /   \      |
                      |   /     \Xfmr  /     \     |  Breaker (w/manual reset)
       Hot ---------------|------------|-----------+--o----o------ Hot
    "LINE"            |   |     |      |     |                       "LOAD"
       Neutral -------+---|------------|--------------o----o------ Neutral
                          \     /      \     / Xcitr    ^
                           \   /        \   /  Xfmr     |
                            ---          ---            |
                             |            ^             |  "Trip"
                             V            |             |  Command
                         +-------+    +-------+         |
                         | Full  |    | Oscil-|         |
                         | Wave  |    | lator |         |
                         | Bridge|    +-------+         |
                         +-------+                      |
                             |                          |
                             V                          |
                         +-------+    +-------+         |
                         | Integ-|    | Compa |         |
                         | rator |--->| rator |---------+
                         |       |    |       |
                         +-------+    +-------+


The heart of the circuit is the differential transformer, through which both
the hot and neutral wires physically pass.  This transformer is simply a
torroidal coil with several hundred or thousand wraps.  As all you EEs know,
if current flows in a wire which passes through such a coil, a current is
produced in the coil windings as well.  The direction of the current in the
wire determines the direction of the current in the transformer windings.  If
you pass equal currents through two wires which pass through the transformer,
the current produced in the transformer windings either doubles or goes to
zero by cancellation, depending on the direction of current flow in the two
wires. You can imagine the current flowing in the hot wire to the right,
through a load on the load side, and back to the left through the neutral
wire.  In this case, the currents in the two wires are equal and the
direction is opposite, so the differential transformer has no output signal
(in theory, see below). 

When someone is getting shocked, current is flowing from the hot wire through
the differential transformer, through the someone and to ground on the load
side.  So some of the current in the hot wire isn't being returned in the
neutral wire, but is instead being shunted to ground through some poor
bastard.  This is a "ground fault".  Since the current in hot and neutral are
nonzero and not equal, the differential transformer starts to produce an
output.  The bigger the fault, the larger the output. 

The signal processing circuitry is familiar to EEs.  The transformer output
is first rectified, then drives an integrator.  The integrator's output
charges up at a rate proportional to the size of the fault.  When the
integrator's output reaches a preset value, the comparator produces an output
which trips the GFI, disconnecting the load side.  Of course, the built-in
receptacle, if present, is wired to the load side of the GFI. In actuality,
the breaker itself is spring loaded with manual reset, and is tripped by a
small pin which is thrust out by a solenoid driven by the comparator
(typically... designs may vary). 

Signal processing folks will recognize that the trip time is proportional to
the magnitude of the fault.  UL publishes a curve of fault current vs. trip
time which the GFI must adhere to.  GFIs must trip with a fault of 5
milliamps, but they have several seconds to do so.  As the fault increases,
the trip time decreases until, at faults on the order of hundreds of
milliamps, the circuit will trip very quickly (30 milliseconds and less is
typical). 

The test button connects a resistor from the load hot to the line neutral.
Typical test currents vary between vendors, but 10 milliamps is not unusual. 

The other transformer ("exciter transformer") is driven by a simple
oscillator to put a small AC waveform through the coil windings.  The purpose
of this is to detect installations which have "grounded neutral", where the
neutral and ground are connected on both the load and line sides.  This
wiring is a problem because the return current now bypasses the differential
transformer and the GFI trips hard and fast as soon as any load is applied.
Grounding the neutral on both the line and load side effectively creates a
wire loop which passes through both the exciter and differential
transformers. In this case, the oscillator output driving the exciter is
coupled to the differential transformer through the wire loop, and the signal
is sufficient to trip the GFI even if there is no load present.  This is done
as an indication that the GFI is improperly installed. 

As far as combined breaker/GFIs, I know of no reason why one should be more
reliable than the other, except for brand-to-brand differences in
reliability.  The industry is pretty mature now and they're all about the
same, to my knowledge.  The biggest problem remains false tripping due to
excessive noise sensitivity.  Another problem is imperfect differential
transformers, in which as the load increases, the transformer starts to
generate an output even though there is no fault.  This brings the GFI closer
to its trip point, decreasing noise immunity. 

								    Marc 

    who-ought-to-know-since-he-used-to-design-these-things-and-the-
    equipment-that-tests-them-for-a-living. 

    RE: .112  


    Whoops! When I said: 

> Apparently, the Code and the makers of GFCI 20A flow/15A outlets  
> don't agree.

    I meant that neither of them agrees with me. I did not mean to say
    that they don't agree with each other.


    Barry    

re .112:

Well, here's my guess as to why the code is less strict here.  
I'm not a code expert, but I have browsed through it.

1)  Homes tend to have a lot of sockets per circuit -- the
code requires lots of sockets.  It would be very expensive
to run separate circuits for each, so it is useful to be
able to have fewer circuits at 20A instead of 15A.

2)  Very few home electrical devices require more than 12A, 
so it would be overkill to mandate 20A sockets for homes.

3)  If a home device can draw enough power to damage the
15A socket, then either it has a 20A plug and can't be
plugged into a 15A socket, or else it has a 15A or less
wire from the plug to the device, and that is going to melt 
or arc before the socket does.


Note that the code has lots of rules to try to keep high
current devices on separate circuits.  For example, I was going
to put sockets in my kitchen and bathroom on the same circuit,
since they are on opposite sides of the same wall, but the code
specifically disallows this -- I guess they don't want my
toaster oven and my space heater on the same circuit.

Also note that the code demands that all hard-wired parts
of a circuit be able to carry the full current, but does
not require that of mundane home plug-in equiment.
Imagine having every wire on every lamp able to carry 20A.
It's a compromise, and the safety level is probably not
as high as for other situations, but that's life.

	Enjoy,
	Larry

    A nit, in the interests of excruciating accuracy:
    
.115>  ...I was going
.115>  to put sockets in my kitchen and bathroom on the same circuit,
.115>  since they are on opposite sides of the same wall, but the code
.115>  specifically disallows this ...
    
    More precisely, if a circuit is shared between the kitchen and some
    other room, that circuit doesn't count as one of the required two 20A
    kitchen small applicance circuits.  A third small appliance circuit, or
    a kitchen lighting circuit, may be shared with one or more other rooms
    if its capacity allows.
    
    Even this nit has an exception:  as mentioned recently elsewhere in
    this conference, a kitchen small appliance circuit may be shared with
    the dining room.

    
    I'd like to ask a mundane question about GFCI applications:
    
    I'm installing outlets on a screened porch.  I'm planning to put a 15A
    breaker in the box, and then a GFCI outlet on the first plug on the
    porch.  The others will be downstream-protected by the GFCI.
    
    Given that it's a screened porch, is it essential that I put
    weather-proof covers on the receptacles?  Or will the GFCI be
    sufficient protection?  I don't think there's going to be a lot of
    water coming in, but.....  If the GFCI keeps tripping in the rain, I'll
    put on covers.
    
    Any thoughts?

    
     rep. 117
    
     Why invite problems?  Put the weather tight cover on! It also might
    be required in your town. 
    
    		Wayne

     All electrical outlets located outside a building and exposed to
    the environment, SHOULD be covered, sealed and calked.
     Sure there is a little extra cost associated with the weather tight
    outlet covers, but, if you don't cover them and they are mounted
    in the body of the house, you can count on rain coming into your
    house.

I want to continue the "stupid electrician tricks" discussion from the
"This Old House" topics (1974.50,+/-10) on GFCIs to this, the GFCI topic.

To recap (so you don't have to retrace the discussion), the electrician on 
This Old House demonstrated the efficacy of a GFCI breaker by dropping
a turned-on trouble light into a bucket of water.  Some of us questioned
whether such a demonstration was reasonable, and whether it actually did
demonstrate the real-life effect of a GFCI.
For example, where did the fault current go that caused the breaker to trip?
The bucket was plastic, sitting on an apparently dry wooden floor.
I observed that the only way the GFCI could have tripped was if the trouble
light had a grounded plug, which most bathroom appliances (radios, hair 
dryers, shavers) don't have.

That got me thinking of other aspects of the ground path, and I asked
whether modern plumbing provides a good enough ground path to cause
a GFCI to trip if the appliance should fall into a sink.  With the
more common use of plastic waste pipes and supply lines, and plastic parts 
in stops and faucets, is a wet sink a good ground path?  
If not, that's both bad and good.  
The good is that if I touch a live connection while also touching
the sink, I don't provide part of a ground path that can endanger me.
On the other hand, if I do drop a shaver into a sink of water, there is
no place for fault current to flow, so the GFCI won't trip as soon as it should.

So, should sinks have a separate hard safety ground?  I observed that
bathroom fixtures in England do.  Is that for just this purpose?

- tom powers]

  With properly working GFCIs, it doesn't matter whether fixtures are
  grounded. The important point is that if enough current returns to
  ground through *any* path other than neutral (especially a path which
  includes a human body), the GFCI will trip. 

  I suspect that the reason for grounding fixtures is that if a live
  appliance falls into a sink, for example, the sink will not become
  live (though any water in it will), and any device protecting the
  circuit may eventually trip. The bad part is that anything in the
  water is also part of the current path, and if the circuit isn't
  protected by a GFI, it won't trip until the current reaches the rating
  of whatever's protecting the circuit. So, as you pointed out,
  ungrounded fixtures are "better" if you're already in the water when
  the appliance falls in. Of course, getting your butt out of an
  ungrounded bathtub with a live shaver in it is going to be tricky. In
  reality, though, even an ungrounded bathtub may conduct enough current
  to ground (through the air and surrounding materials) to be lethal. So
  since it may still kill you if you're in the water, might as well at
  least save you if you're lucky enough not to be in the water when the
  appliance falls in. I guess that's a good enough reason for ALL
  plumbing fixtures to be grounded.

  In the absence of GFI protection, a sink may be safer grounded
  (because you tend not to be in the water), while a bathtub may be
  MARGINALLY safer floating (because you're probably in the water when
  the appliance falls in). With GFI protection, it doesn't make much
  difference.

  One thing I've been curious about is to see if a GFI socket will trip
  when I stick one of those neon test screwdrivers firmly in it. I've
  noticed (by accident) that the screwdriver I have can give a moderate
  shock if I hold its metal end firmly in one hand and grab a grounded
  object in the other (the neon bulb also lights up nicely). I haven't
  tried this on a GFI socket, yet. Has anyone out there been saved by a
  GFCI?

You cannot rely on plumbing fixtures to have a good ground unless you know that
they are solidly connected with copper pipe all the way back to the house
water inlet.  But it really doesn't matter.  A GFCI detects leakage as
a difference in current flow between the hot and neutral lines.  It doesn't
matter WHERE that leakage goes, as long as it is going somewhere else.

				Steve

re-.1 ...copper pipe all the way..
I recently had the local cable company out installing cable into my house
while doing the inside work hesaid he had to ground to the CW pipe.
I told him that I had replaced it with plastic all the way to where it entered
the house except for the few feet where he intended to connect( this was partly 
in an effort get him to use the ground rod inches away from where the cable came
into the house as I believe grounding causes premature pipe failure) anyway,
he said " oh well I can't help that it's company policy to use the pipe...".
Oh boy and they let these guys climb power poles.....
The guy was still parked in the driveway packing up when I moved the ground
wire to the rod where it could do some good.

-j

> < Note 1659.122 by QUARK::LIONEL "Free advice is worth every cent" >
> ...But it really doesn't matter.  A GFCI detects leakage as
> a difference in current flow between the hot and neutral lines.  It doesn't
> matter WHERE that leakage goes, as long as it is going somewhere else.

Exactly my point - if the sink is not well grounded, or not grounded at all,
where can the current go to cause the imbalance in hot and return?
Should sinks be grounded separately?  Do either the plumbing or electrical
codes have statements on this?

- tom]

side note: Clean water and dry wood are not very good conductors of electricity.
Wet wood is better.  Wet skin - now THERE'S a good conductor.

Re: .12

Water is a good enough conductor, clean or otherwise, that the few milliamps
needed to stop your heart are very likely to find a path to ground,
whether the sink itself is grounded or not.  I don't see any particular
point in ensuring a sink is grounded, but I would not hesitate to
protect any outlets near a water source with a GFCI.

					Steve

Pure (distilled) water is a poor conductor.  Water with the usual impurities
is a fairly good conductor.  It is highly unlikely that you have distilled water
in your plumbing, so it's safe to assume any water in pipes, sinks, etc.
is a good conductor.

-Mike

    re: grounding a sink.

    How would one ground a plastic sink (or fiberglass) with a plastic
    drain pipe? (This is a question of curiosity, not need). 
    
    -Bob

Re: .12

You wouldn't, any more than you'd ground a toilet.

		Steve

To expand a bit on the note a couple back -- it doesn't take many milliamps
to kill you because the way it kills you is by interfering with the neural
signals to your heart and starting "fibrillation", which is a sort of
random twitching.  Once the heart gets stuck in fibrillation, it is hard
to get it back to regular beating, and you haven't got long to do it.

A stronger current will instead clamp your heart, and if you get free before 
you fry, you have a fair chance that your heart will start beating again by 
itself.  That's one reason why people have survived lightning bolts that
melted the coins in their pockets.

	Enjoy,
	Larry

re < Note 1659.129 by RGB::SEILER "Larry Seiler" >
>                     -< why relatively low currents kill >-

>A stronger current will instead clamp your heart, and if you get free before 
>you fry, you have a fair chance that your heart will start beating again by 

	So, what Larry is saying is, if you find yourself getting zapped
	by something, grab hold of it tighter, or toss some table salt
	in the water, to increase the current so your heart clamps up,
	until someone finds you and turns off the juice!

	(many many :-) :-) :-)...!!!!)   Josh

It also depends upon the frequency of the current.

I once got attached across a 400V DC power supply.  Before I was
disconnected, I was drawing 40ma (from one hand to the other).  At DC, 10ma
and you can't let go, 100ma and you're dead.  (some of my friends claim that
I therefor became 40% dead, explaining many things about my behaviour :-))
The current required to freeze or kill goes down with frequency, with a
minimum at about 47Hz.  The 60Hz used in the US is remarkably close to this
minimum.  The ring frequency used in telephone systems is even closer (I
seem to recall about 45Hz).

The sensitivity to frequency is due to the fibrillation, as mentioned in
.129.

	-JimJ

RE: "At DC, 10ma and you can't let go, 100ma and you're dead."
    
    Wait a minute, I thought the argument between Edison and Westinghouse
    concerning DC vs AC included something about AC could kill you more
    easily than DC.  100ma DC ain't that hard to come by.
    
    'course this doesn't matter to the GFCI discussion underway, ...
    
    ed

    RE: .132, that's 100ma flow, which must take into account skin
    resistance, which is normally quite high. If the skin is wet though, it
    drops considerably. Also, the path of the current flow is critical. A
    zap from the thumb to the pinkey on the same hand, though painful,
    probably won't be lethal, but the same zap from the left hand to the
    right hand (across the heart) could be deadly.
    
    Now back to our regularly scheduled topic........
    
    Eric
    

    Personally, I was curious about how well GFCI's work and did a little
    experiment late one night.
    
    I took an ordinary detachable power cable (you know the kind we use for
    all sorts of terminals and PCs) and stuck a paper clip into the lead
    I wanted to test.   When I touched the neutral to a steel sink, the
    outlet reset nicely.  When I touched the hot to the sink, it MELTED
    the paper clip at the point of contact before the outlet reset.
    
    I'm still not sure what to think of this.....
    
    	Dave.

    1/40 second is the figure I've read.  They're not intended to
    protect against a direct short.
    
    			Steve

  It is an interesting question, though. When you press the TEST button
  on a GFI receptacle, what does it actually test? The delay before the
  circuit opens is considerably longer than 1/40 sec (25 ms). It appears
  (to me, at least) to be more like 150 to 200 ms. I suppose that this
  is still enough time to protect against serious injury, though it
  would hurt like heck. The GFCI breaker on my panel trips almost
  instantly, though.

  I've been considering a similar experiment: measure my skin resistance
  (left hand to right hand) with a VOM (it usually runs a around 100K
  ohms) and stick the equivalent resistor in a GFI socket to see if/when
  it trips. If I remember to try it, I'll let you know what happens.

    re: .136

> It is an interesting question, though. When you press the TEST button
> on a GFI receptacle, what does it actually test? 

    Having taken a GFI receptacle apart, the one I had (don't remember the
    manufacturer) actually shorted the ground lead to hot (black).  I had
    wondered the same thing as to what was really tested.  I now believe
    the test.

    I too have noticed that the delay seems rather long.  Who knows, maybe
    it has AI and "knows" it's only the test button being pressed so it
    doesn't have to react all that quickly. :-)

    -Bob

  Hmmm, ground-to-hot doesn't strike me as right, because if I short
  ground to hot on a normal socket, the breaker for that circuit will
  trip in a couple of milliseconds -- long before the GFI action. I had
  therefore surmised that it must be ground-to-hot through some
  resistance which wouldn't trip the branch circuit breaker. Can anyone
  confirm this?

I believe they connect hot to ground through a resistor that draws enough
current so that it should trip, but not much more.  This is how I'd do it
if I were designing such a thing.

If the test button connected hot to ground directly, you'd know it, because
there's be a loud BANG that would scare the %#^$%^#@ out of you.

-Mike (who has accidently connected hot to ground more than once when playing
with old radios as a kid)

>            <<< Note 1659.136 by VIA::GLANTZ "Mike, DTN 381-1253" >>>

>  I've been considering a similar experiment: measure my skin resistance
>  (left hand to right hand) with a VOM (it usually runs a around 100K
>  ohms) and stick the equivalent resistor in a GFI socket to see if/when
>  it trips. If I remember to try it, I'll let you know what happens.

This won't be a very good test.  People are non-linear conductors.
The salts and electrolytes insode our bodies are good conductors, dry skin
is not so good.  Note the 100K or so you see is pretty much constant
regardless of the distance between the leads, as left hand to right,
or both in the same hand.
Add a little bit of sweat, and the resistance drops fast.  Moisten
your skin and try it, or put the VOM leads on your tongue.
When you touch a voltage source, you'll draw a bit of current, which might
squeeze a few sweat glands, causing you to draw a lot more current very
quickly.

- tom]

  Of course you're right. I was just looking for a rough test to see how
  GFCIs work. I know it will trip at a low resistance, but I'd like to
  see it trip at the higher resistance -- it would make me feel safer.

  You're probably also aware that skin resistance varies a lot from
  person to person, as well as in a single person under different
  conditions. I can touch a 110VAC hot (with dry skin) and feel minor
  pain. My wife hits the ceiling. An electrician friend can touch both
  neutral and hot simultaneously with any parts of his body and feel it
  but not be bothered at all. I even watched him put his fingers across
  the two hot bus bars in my breaker panel and not complain (and not
  trip the 100A main breaker :-).

    Hmmm..  ground to anything does not seem right, because as I understand
    it GFIC's work on *ungrounded* outlets as well as grounded ones.  So if
    a ground is not connected how can you short to it.
    
    My guess is that the test button somehow simulates  a current imbalance
    between hot and neutral without connecting anything to ground.
    
    gjd

  That's an interesting point. I just might go home and try this tonight
  by lifting the ground on a GFI receptacle.

  Meanwhile, I can't imagine any other way to test GFI protection (aside
  from sticking your finger in the hot side of the socket), since in
  order to trip it, you need to cause a difference in current flow
  between neutral and hot, and in order to do that, you need to send
  enough current from either of these to some other place. Without a
  ground path, what other place is there?

    hmmm.... I don't know for sure, but I don't think it's legit to 
    install a GFCI on an ungrounded circuit.   It may not work correctly.
    Can someone verify this?  (both functional and code?)
    
    	Dave.

  No, in fact the National Electrical Code specifically prescribes GFI
  receptacles to be installed on existing ungrounded (two-wire)
  circuits, rather than installing a standard (three-wire) receptacle
  which couldn't be grounded. You can either do that or install a
  two-wire receptacle (the old-fashioned kind). The GFI receptacle still
  provides protection, and doesn't need a ground to work. I have no idea
  how the test button works in that sort of installation, though.

    The 1990 NEC now allows you to feed other outlets from the GFCI
    using two-conductor wire.  I have made use of this in my house, and
    yes, the test button works.  Don't ask me how....
    
    				Steve

The test button probably works by connecting the hot lead after the current
sensor, through a resistor (to limit current flow) to the neutral lead _before_
the current sensor.  This way the current flows one way through the current
sensor but not back the other way, so there is a current imbalance, so the gfi
trips.

If you think of this, this is exactly the same as using the ground lead for the
test button, since the neutral and ground are connected directly together at the
main panel anyway (the electricity doesn't care about human laws regarding
where to connect the neutral and ground)

-Mike

  Bingo. I'm sure you're right.

    Yes, and the bias is calibrated for 5 millamperes.
    
    Erick

WHen I bought my house, the owners mentioned something about the outdoor outlet
and the bathroom, which I promptly forgot, until recently.

I recently hooked up a bug zapper to my outdoor outlet, and after a few weeks
of constant zapping (I own the neighborhood swamp, ya know), it quit.  Then it
started working again.  After a few times, I finally figured out that my the
times my electric toothbrush was discharged in the morning due to the GFCI being
tripped coincided with the nights that the zapper didn't work.  Some testing
confirmed that the outside outlet is somehow wired to the GFCI in my bathroom,
and is controlled by the tripping of the reset switch.  

Aside from the bizzarity of the circuit connection, is this unusual?  Can
GFCIs control other devices?  Does this mean my outdoor outlet is protected the
same as my bathroom outlet?

    You can string a whole bunch of other outlets from a GFCI
    outlet; this is typically done by CHEAP electricians who
    use a single GFCI breaker/outlet for ALL bathrooms and outdoor
    outlets.  All following (fed-through) outlets are then GFI
    protected.
    
    Wouldn't be at all suprised if you had water in your outside
    outlet box; maybe you could correlate trip times with whether
    or not you had rain?  The bug zapper could have an insulation
    problem of some sort as well.
                                    
    Then again, maybe you just have too much of a load on that
    (supposed) single circuit.

    
    Hi,
    
    	The ballast for the U/V lamps are also a good source of problems
    for GFCI trips as are fluorescent fixtures, same principle of
    operation. They won't all do it, but a large enough portion of them
    will, that I have been avoiding putting them on GFI circuits when
    avoidable. 
    
    	My suggestion is to have a circuit run for the bug light that
    is not GFI protected. 
    
    PL
    
    

    >>	My suggestion is to have a circuit run for the bug light that
    >>is not GFI protected. 
    
    You can't do that by code, outside outlets have to be GFI'ed, except, I
    believe, when they are used to control wheelchair lifts (its a long
    story....).
    
    
    bjm

In my previous house, the outside outlet was controlled by the GFCI in the
bathroom.  Confused the heck out of me when I went to replace the outside
outlet with a GFCI and found it didn't work.

But there was an up-side.  If I wanted to disable the outside outlet, all
I had to do was trip the bathroom GFCI!

				Steve

Well... this might explain a few things.  For one, I intermittently get one or
both lights lit on the bug zapper for no apparent reason.  And, as far as too
much load, I think the zapper ran non-stop while it was dark (I mean in bug-fry
mode, not just turned on) for a couple of weeks.  I mean the dead bug pile had
DEPTH!  On the down side, the neighborhood ants love the crispy critters so 
much they moved in directly under the zapper, and have tried to carry off my
dog.

I'll try a few experiments with pwering the thing from non-gfci outlets, to 
see what happens.

Bug zappers supposedly aren't very good at attracting mosquitoes, since
they're attracted by carbon dioxide, not light.  Most of the bugs you're
frying are either good bugs or so-so bugs.

Living at the swamp, I get lots of bugs.  Good, bad or otherwise, they are soo
thick, I can't get in or out of the house in the evening without letting in
a few dozen.  My neighbor, a firm believer in zappers, gave me this unit when
he bought his new one (auto on/off).  Since I've had it on, there has been a
marked decrease in the amount of bugs in the house, the amount of bugs near our 
doors, and the amount of fried bugs piling up on the ground below.  Also, I'm
the mosquito attractor in the family.  People invite me to parties and ask me
to sit at the end of the yard to keep all the mosquitos away from everyone else.
I get bitten mid day on hot dry days.  Each bite I get swells incredibly and
itches a lot.  I've always been too skeptical of zappers to go out and buy one,
and I don't like to spray insecticides, and I'm not allowed to have bat houses.
However, this deal was too much to pass up.  And so far it works!

But, I digress...this is after all the GFCI note.  So,...

I found out the source of all my problems is the GFCI-ness of the outlet.

Is there any way of getting around the code restriction on outdoor outlets?
Like direct wiring? (or violating code? I wouldn't do it, but I would like
to know if it's the ONLY way...)

re .157:

> Living at the swamp, I get lots of bugs.

How serious are you?  If you live near the Okefenokee, you're out of luck.
If you have a fairly small amount of stagnant water, you might try
Bt Israelensis, a bacterium that attacks mosquito larvae and is harmless
to other species.

> I'm not allowed to have bat houses.

Are you allowed to have purple martin houses?

> But, I digress...this is after all the GFCI note.  So,...

Oops, so do I.  If your GFCI outlet is tripping, there's a problem in
something you've got plugged into it, no?  Why ignore the problem?

  You won't be easily able to eliminate the GFCI on the zapper circuit
  without violating code. It would be better to try to figure out why
  the zapper is tripping the GFCI.

  GFCI receptacles are specifically *not* designed to trip in
  overcurrent conditions. The conventional overcurrent protection on the
  circuit (fuse or breaker at service panel) is intended to handle this
  condition. The GFCI only trips for leakage completely out of the
  circuit (and, presumably, into a human body). This is probably what's
  happening in your case.

  You need to find out where/why current is leaking out of the circuit.
  This may be inherent in the nature of bug-zapper operation. Any
  high-voltage appliance will leak some current into the surrounding
  environment, especially when it's humid. If this is true, then there
  are going to be a lot of people with the same problem, as more and
  more homes get built or upgraded with GFCIs under existing code.

  There are two possible solutions: either find a zapper which doesn't
  leak enough current to trip the GFCI (maybe newer models are better in
  this regard), or find a way, compatible with code, to install the
  zapper on a non-GFCI-protected circuit. I seem to remember some
  exceptions in the code (for dedicated appliances) which would allow
  you to do this. Maybe somebody with their codebook handy could confirm
  this.

  BTW, the GFCI may be tripping for some other reason having nothing to
  do with the zapper. That would be easy to fix.

    
    Hi,
    
    	There are ways to have the bug light not on a ground fault device.
    
      *	1) hardwire the buglight to the circuit....no GFI needed then.
           
    		I would choose #1 
    
    	2) a little looser interpretation of NEC GFI requirements is
    	   to mount the receptacle higher than 6 foot 6 inches above
    	   grade making sure that the receptacle cannot be reached from
           the grade around the receptacle.
    
    	3) Change the ballast to a 230 volt ballast, convert the receptacle
    	   to the correct NEMA configuration for the voltage and the
   	   amperage available, add a 2 pole breaker to the circuit.
                                                                   
        4) You might also try another ballast, you may get lucky and
     	   get one that works with a GFI.
    
       (Refer to Article 210-8 , paragraph A, section 3, NFPA 70, 
    	1990 National Electrical Code for the specifics.)
     
    		The purpose of GFI protection is to protect persons
    	from being electrocuted while using portable equipment that
        is cord and plug connected while in contact with the earth. 
    		
    		If you take away the ability to use cord and plug connected 
    	equipment, or remove access to the earth while using the cord 
    	and plug connected equipment ie, grade level out of reach such as
    	a second floor deck with no stairs to the grade level or by
    	placing the receptacle out of reach ie more than 6'6" above
    	grade level (NEC) or use a branch circuit that is greater than
    	125 volts and or greater in amperage than 20 amps, you will find 
    	that in most cases you will be complying with NEC....but once
    	again there are exceptions to every rule and local variations 
    	also exist.
    
      		The above only applies to residential, there are other
    	ground fault rules that apply to commercial and industrial
    	installs.
                
    		Hope this helps some,
    
    	   

    
    	RE: reply .160
    
    		Mike is absolutely correct, check out the circuit with
    	the Bug light unplugged to be sure that it isn't something else
    	causing the trip.

    I was thinking along the same lines as .161, hardwire it.  That means
    taking it apart and replacing the cord so it is wired directly into 
    a weatherproof box.
    
    Then I was thinking that if you changed the male end of the plug to a
    non-standard configuration (such as a twist-lock) you would eleminate
    the issue of having anything else except the bug light pluged into this
    receptical.
    
    The problem with these solutions is that neither one of them will save
    your life if the bug light falls into the pool of water you happen to
    be standing in (or if the metal pole you are cleaning a pool with
    contacts the zapper in such a way to electrify (sp?) it).  My final
    solution would be to see if the GFCI receptical was faulty and then if
    that did not work try to fix the bug light.
    
    Brian

Well, I've never had a problem with the GFCI tripping 'till now.  But, I've
only used the outside outlet for minor short term use (weed whacker) with 
ungrounded (2 prong) plugs.  The outdoor outlet does look a bit worse for the 
wear, so I'll replace it and go from there.  

While I'm here, if I wanted to direct wire the unit, what kind of wire is needed
between the house and the tree where it hangs.  Is conduit needed?

Another thing you may want to try if you have another GFCI outlet in your 
house is to get an extension cord and try to plug the bug zapper into the
other GFCI.  I have had faulty GFCI's before, sometimes they are overly
temperamental and trip their breaker at even very low circuit fluctuations.

If you find that this is the case, then just replace the GFCI with a new
one.

Chris

re: .164

	Some other things that you can do is ensure that your not getting
water into the extension cord connection. I.E. It will short at the juction
of the plug and the extension cord. I would check and ensure its dry and
then tape over this joint.


re: .165

	Could be your problem too. (A defective GFCI)


Bill

    
    Hi again,
    
    		The Bug light is in the tree with an extension cord
    running back to the house outside outlet? Had any rain showers lately?
    	Those flip up covers are not water proof with the plug plugged
    in and the cover flipped up. For permanent installations a different
    type of outlet cover is required, one that stays waterproof with
    the plug plugged into the receptacle. 
    
    	Is the GFI a breaker type or receptacle type? If receptacle
    type is it located outside in that receptacle box? I have also
    had problems with GFI's located out of doors in boxes like that.
    They get damp and trip. Dampness in the box outside can also trip
    GFI's if they are located on the inside of the house too.
    
    	Isn't troubleshooting fun over the network :')
    
    

re .161:

The folks who wrote the code must be very short if they think you can't
stand on the ground and plug something into an outlet 6'6" above grade.

Well, another weekend went by without doing much about it but looking.  

The box outside is not a GFCI outlet itself, but us wired to a GFCI outlet 
(not breaker) inside.  The out door outlet could be having moisture problems.
It is a flip up style outlet, and the plastic surrounding the holes is cracked 
in a few places.  My first step is going to be to replace this outlet.  I am
curious about the outlets that remain weatherproof when something is plagged in.
Is there a name for them?  Cost?  Place to buy?  The existing outdoor outlet
is flush with the house.  An outlet whose box is surface mounted would interfere
with traffic up and down the back steps.  Are these weatherproof outlets 
available in flush mount (or at least such that only the plug and cord stick out
above the surface?  

					Thanks,
						JP


ps: I did tape the plug into the extension cord with too much black plastic
    tape.  I did resist putting a layer of duct tape over that, however.  
    (almost any problem can be solved with duct tape...)

    
    Hi again,
    
    		Try getting a HUBBELL brand wiring device catalog. In
    it you will find a section on marine/marina wiring devices...not
    cheap but they work well and would satisfy the application you appear
    to have. These catalogs can be had for free from your local electrical
    supply if they are a HUBBELL dealer. There are other manufacturers
    of this type of equipment, BRYANT, EAGLE, happen to come to mind
    but my experience has been best with HUBBELL.
    
    

>>		Try getting a HUBBELL brand wiring device catalog. In
>>    it you will find a section on marine/marina wiring devices...not
>>    cheap but they work well and would satisfy the application you appear
    
    They do may good stuff but THEY ARE NOWHERE NEAR CHEAP.  It is not just
    their marine stuff.  Anything they make is very good quality and very
    high priced.

    
    	You are correct they are not cheap......I never said they were.
     
    I have been hearing the same comments about certain brands of
    electrical equipment (hubbel, square-d) for the fiften years I have 
    been in the trade. All I can say is you get what you pay for.

Any Master Electricians out there have an opinion on this?  I can't
find anything in the NEC which describes this situation.

I want to install double-ganged duplex receptacles above my workbench
to increase the number of outlets available without using up lots of
wall space, feeding each of the two outlets in the double-gang box from
a separate branch circuit to increase the total ampacity available.
In the "pre-GFCI" days this would be easy; run 14-3 from a two-pole
breaker, observe the NEC requirements about integrity of the grounded
conductor, and daisy-chain the 14-3 to subsequent boxes (a total of
three boxes in the chain => six duplex receptacles).

I want to use a similar scheme in my new house, but intend to use GFCI
receptacles at the "head" end of the chain.  This means I need to run
two separate 14-2 cables to the subsequent boxes (in which are mounted
standard duplex receptacles).  This leads to an interesting question.

In all the subsequent boxes, the grounded conductors (white wires) of
the feed-throughs must remain distinct, otherwise the head-end GFCIs
may become confused due to current imbalances.  Does the NEC prohibit
such a distinction?  Normally one would expect all grounded conductors
(white wires) to be tied together (discounting the case of "identified"
white wires used for switch loops and etc.).

I don't know about the GFCI's, but for a workbench outlet you should definitely
use a 20A circuit, not 15A.  Many hand power tools will be too much for a 15A
circuit, particularly at startup.  And for the 20A circuit, you'll need 12-2 
wire.

Paul

    RE:0

    	You are correct. One must keep the grounded conductors separate as 
    soon as you leave the GFI's. But the assumption that all grounded
    conductors are typically tied together is wrong. If you have 2
    separate circuit feeds in a box, one must keep the grounded conductors
    separate. Just as you would keep the ungrounded conductors separate.
    The only conductors that get tied together in a box regardless of the
    circuit that they belong to is the grounding conductors. This ensures
    that there is a zero voltage potential between grounding conductors at
    each location.


    			CB 
    

re .2:
	I had to read this a few times.  Using the correct terms grounding and
grounded may confuse people.  If I may:

>>    separate circuit feeds in a box, one must keep the grounded conductors
>>    separate. Just as you would keep the ungrounded conductors separate.

    separate circuit feeds in a box, one must keep the grounded (neutral) 
    conductors separate. Just as you would keep the ungrounded conductors (hot)
    separate.

>>    The only conductors that get tied together in a box regardless of the
>>    circuit that they belong to is the grounding conductors. This ensures

    The only conductors that get tied together in a box regardless of the
    circuit that they belong to is the grounding (usually the bare copper, or
    green insulated/marked) conductors. This ensures

	I also agree with .1, use 12 guage wire and 20 amp breakers.  The
price difference is little and the feeling of security is great :-)  You can
use the same GFCI outlets and chain them.  Just make sure they are rated for 15
Amp and 20 amp through.

	I assme you are going with GFCI's because of the change in the NEC 
about outlets in an unfinished basement?  Just what does constitute an
unfinished living space anyway?

Brian

    Re .2:  Thanks for the verification!
    
    Re .3:  About the only nit I'd pick with your clarification is the
    use of the term "neutral" for the grounded conductor.  You can only
    have a "neutral" conductor in a 3-wire 120/240 volt circuit; once
    a 2-wire 120 volt circuit branches from the panel, technically you
    no longer have a neutral conductor, but rather a grounded conductor.
    
    You're right, though, that the "ed" and "ing" suffixes can be a
    little confusing, but if you're going to do electrical work you
    really should learn and understand the differences between the
    "grounded" and "grounding" terms.
    
    
    Yes, 12 gauge is what I will be using (although I might jump to
    10 gauge for a long run to minimize voltage drop).  I'm using
    GFCIs primarily because of the safety factor; I'd probably use
    them even if the NEC didn't require them.

I'd like to know what outlets are downstream from a GFCI outlet.  If I hit
the test button, will all downstream outlets go dead?

    Yes, assuming the GFCI works properly and the test button really
    does trip the interrupter in the outlet.

In .0, 

> In the "pre-GFCI" days this would be easy; run 14-3 from a
> two-pole breaker, observe the NEC requirements about integrity of the
> grounded conductor, and daisy-chain the 14-3 to subsequent boxes (a
> total of three boxes in the chain => six duplex receptacles).

Is three the maximium number of boxes allowed in a chain? What's
the maximium number of boxes in total?  I'm in a house built in 
1979 and I think there are circuits with more than 3 boxes.
Of course, the code may have changed since then.

Regarding .3

> I also agree with .1, use 12 guage wire and 20 amp breakers.  The price
> difference is little and the feeling of security is great :-) You can
> use the same GFCI outlets and chain them.  Just make sure they are rated
> for 15 Amp and 20 amp through.

This in not intuitive. I always thought that you needed 20 amp devices
for 20 amp service. Aren't you required to have 20 amp outlets on
20 circuits? Of course few people do.

    No, the "three boxes" was just in my example.  The last time
    I checked the NEC there was no explicit limit on the number
    of receptacles on a circuit, but most inspectors use the "rule
    of thumb" which says allow 1.5A per receptacle.  This gives a
    limit of 8 per 15A circuit, 12 per 20A circuit.  (This is for
    120 volt outlets; 240 volt is another matter.)
    
    Regarding the rating of devices on 20A circuits, what is really
    necessary is that the device is rated for 20A feed-through
    capability (most 15A receptacles ARE 20A feed-through rated).
    This simply means that the connections on the device are such
    that they can handle a 20A current.  The NEC allows you to
    connect such a 15A receptacle to a 20A circuit.  It also will
    allow a 20A receptacle "downstream" of the 15A one.
    
    Most GFCI receptacles have a 20A feed-through rating.
    
    (Note that a 20A plug has the groundED blade at a 90 degree angle
    to the normal orientation.)
    

    I wish my problem had *remained* that simple :-(  My wife opened the
    garage door but could not close it.  I found that it was not receiving
    any power.  Eventually I traced it GFI ckt in the upstairs bathroom.
    It was NOT tripped but TESTing and RESETing it seemed to fixed it, at
    least for a while.
    
    However, when I tried to explain and show it to my wife, I discovered
    that once I again I have lost all the ckts connected to that GFI.  I
    have repeatedly TESTed and RESETed it with no effect.  I have also reset
    the corresponding circuit breaker at the main pannel, just in case.
    
    I borrowed a small plug tester from the neigbor.  LED display shows
    something like "Hot/Neut Rev". If I trip it with TEST button, no LEDs
    light up.
    
    Any advice?
    
    Thanks,
    - Vikas
    
    P.S. It is relatively new (1986) construction.

Re: .13

If the tester showed "hot-neutral reversed" you have a miswired outlet and
a VERY dangerous situation.  That might also explain why the GFI is not
working properly.  Turn off the circuit breaker for that circuit and
rewire the GFI correctly - black wire should go to dark-brass colored
terminal (or black wire), white wire to silver-colored terminal (or
white wire).  The tester should show two yellow lights and no red lights
when plugged in.

				Steve

    All the wires go to the right places and the screws are tight.  I
    refused to touch anything if the ck is live.  Herre are the results
    with breaker off :-
    
    1. GFCI tripped (I needed to turn the breaker on before I could trip
    		the GFCI, so at least it does create a fault to test it!)
    
    	No Continuity between Load Hot and Line Hot 
    	No Continuity between Load Neut and Line Neut
    
    2. GFCI RESET
    
    	Continuity between Load Hot and Line Hot 
    	*BUT* No Continuity between Load Neut and Line Neut
    
    It does look like that the GFCI itself has gone bad.  This looks like a
    Leviton product.  Does it come in standard size?
    
    - Vikas
    

GFIs do go bad.  They are standard size, and the Leviton model is the most
widely available and is pretty cheap - prices as low as $7.50 have been
seen lately.

					Steve

        If you are talking about the "standard" outlet tester,  it won't
        tell you anything unless the circuit is LIVE. If you suspect
        problems and are afraid to plug it in live, then first plug it in,
        then turn the breaker back on. Anything other than the two green
        lights ON and the red light OFF is a problem.
        
        If that tests OK and the GFI still trips, then either the GFI is
        bad, or you have a wiring problem or bad device plugged in
        downline of the GFI. With the breaker again off, disconnect the
        downstream leg from the GFI to tell if it is the GFI itself is the
        problem, or if it is the downstream leg.

And also, when a 3 prong circuit tester says your circuit is bad, it isn't
always bad in exactly the way the thing says.  It depends on how creatively
your circuit was wired!  For example, if both the hot and neutral are at
high voltage, a 3 prong tester will tell you that neutral and ground are
reversed.  I had to get a 2 prong glow tester to figure that one out.

	Enjoy,
	Larry

I had a similar question when I was looking for an ITE 240v 50A breaker when I
installed a spa.  The folks at Ralph Pill told me that such a beast did not
exist.  GFI 240v breakers went to about 30A, and that was it.  (Take this
with a grain of salt, however, I didn't persue it very far, so such a beast
may really exist.)

There are two solutions that come to mind:

   1)	Create your own ganged 240v breaker by bolting the handles if two
	120v breakers together.  (I'm not sure if NEC code would allow this,
	I suspect it may not.)  I don't know how 120v go, as far as amperage.

   2)	Build your own GFI circuit.  I have a book with at least one GFI
	circuit in it.

- Mark

    I checked with Seaman's Supply here in Nashua and was told that ITE
    doesn't go above 30A for a 240V GFI breaker - however, Square D does
    make such a beast.  You COULD use a standard ITE 2-pole (240V)
    breaker and feed a Square D subpanel which would contain the Square
    D GFI breaker.  A little more expensive, but at least you get the
    GFI protection for which you're looking.
    
    Bolting two single-pole GFI breakers together wouldn't work - since
    there would be NO current through the white (groundED) conductor, the
    breakers would immediately trip under any load.

    	Bruce,

    	The NEC does not require GFCI protection on 240v receptacles. The
    intent of the law is to protect the fool with the 2 wire (or 3 wire
    w/broken grounding conductor pin) extension cord, or cords on tools. If
    the tool at the other end has a "ground fault" and there is no ground
    in the cord, the GFCI will see the current imbalance and trip. The same
    situation without GFCI protection, the fault current would take the
    lowest resistance path to ground (this is probably the person using the
    tool). Most 240v tools, or those of 30amp and up, have properly
    grounded cords, and the cannot plug into standard cords (the big
    problem). Thus, the NEC doesn't require protection on "all" outside
    receptacles. What would be within the law (although, I will recommend
    something else), is to install the receptacle necessary in a weather
    proof enclosure and cover. "I" would install the receptacle inside,
    where you could run a longer cord to it (this will eliminate allot of
    corrosion that will occur if it is installed outside). As long as your
    cord is in good shape, and all of the connections are tight, you
    shouldn't have a problem. This is what the grounding conductor is there
    for. It may not sense a small leakage to ground, but it will provide a
    "lower" resistance path (as opposed to you) to ground. And it "will"
    trip on a dead short to ground.

    	CB

I have several comments on the situation.

First, ganging breakers, GFI or not, is not recommended unless specifically 
permitted by the manufacturer.  Most (all?) ganged breakers have internal
common trip hardware in addition to the ganged handle, partly to assure trips
regardless of handle position. 

ITE does have provision for special GFI situations.  You need to directly 
contact ITE and tell them your application, and they will tell you the parts, 
e.g. current transformer and breaker, needed for your situation.  We have done 
this for some of our lab three phase applications.

I doubt if you will be able to use a GFI on a welder, mostly for noise 
reasons.  GFIs sensitive enough to prevent shock are often fairly sensitive to 
electrical transients.

The other welder issue, is what are you trying to protect against.  Arc 
welders come in two varieties, the first has an isolating transformer which 
separates the supply from the welding electrodes.  There is no conductive path 
from the supply to the electrodes, only magnetic coupling, meaning no path to
trip the GFI.  For magnetic reasons, the primary winding is separated from the 
secondary, usually by a couple inches, so primary to secondary shorts are 
unlikely.  In this case, the only thing you may be protecting against is a
short from the supply wiring to the case of the welder, a situation which is
better controlled by good grounding of the equipment. 

If you have one of the cheap arc welders which simply uses a choke in series 
with the phase conductor, any connection of the work electrode (neutral) to
a grounded object will trip the GFI.

Of course there is always the shock potential between the welding electrodes.

RE: Home grown ganged breakers

    I agree that trying to roll your own ganged breaker is a waste of time.
    (I suspected there were a number of reasons not to do it, code violation
    not the least, but didn't have a good grasp on what they were.)

It is nice to know that GFI breakers are available 240v at higher amperages.
Maybe I'll never need one, but in case I do...

Now if only GFI breakers weren't so darned expensive :-(

- Mark

    Here's one that is puzzling my father and me:
    
    GFCI in a metal box, fastened to the bottom of the breaker cabinet.
    3 wires run from the various breaker terminals, to "LINE" and ground
    of the GFCI.  3 more run from "LOAD" and ground (hm, I think that was
    true) to go off to bathroom outlets.  Worked fine, but the positioning
    of the box on the wall was deemed awkward.
    
    Box moved from chest-high to about 16" off the floor;  wires cut and
    run;  GFCI hooked up -- backwards (hot & neutral reversed, smooth
    move!) -- and it tripped.  Hooked up correctly, and it tripped when
    power was applied.
    
    Suggested that perhaps hooking it up with hot and neutral fried
    something in the GFCI, and it would have to be replaced.  So it has
    been replaced -- and also the wires running to and from it, and it
    still trips.
    
    It appears that the first outlet downstream from the GFCI is one
    mounted on the exterior of the house, FWIW.
    
    Can anyone offer any suggestions?
    
    Dick

I wouldn't think swapping hot and neutral would damage a GFI outlet.  Maybe
a corroded contact on the neutral when activated with 110V AC became a
leaky circuit to ground or something.

Try disconnecting all "load" wires and see if it still trips.  If so, it is bad
after all.

-Mike

    With no load, it works;  but that leaves the downstream outlets dead.
    
    Dick

I suppose you could go to each outlet in line and disconnect the downstream
side until either 1) you find where the problem is or 2) you get to the last
outlet without finding it, in which case it now works :-)

-Mike

    Does it trip with no circuit and does the test button work? If not it's
    hooked up wrong, you have a bad GF breaker, or you have a bad ground.
    If that's ok then continue reading.
    
    Hook just an outlet (or another simple circuit in the house) to the new
    ground fault circuit and test it. It should trip when you put a load
    between "Hot" (black side with short slot) and Ground. It should also
    trip from "neutral" (white side with long slot) to ground. And of
    course it should not trip when you put the load between "hot" and
    "neutral". You can't use a multi-meter or neon light for a load test. A
    coil tester is easiest but you could use a lamp with homemade probes.
    
    If that works then It could be that the old GF was not as sensitive,
    not working, or not installed properly. AND that there is a fault in
    the circuit. It is very common to have a fault in an outside outlet.
    They tend to leak water into them. The fault is very likely between
    "neutral" and ground. You'll just have to take each box on the line
    apart as you go down the line to find it. Just taking the box apart
    will usually cause the fault to go away. Putting the box back together
    properly is the trick to doing it right (ground lead does not touch
    neutral terminal on outlet).
    
    Is this the outlet style or breaker style GF?
    
    My GUESS is that hooking the GF backwards would not blow it and perhaps
    when you hooked up either GF correctly your detecting a fault that has
    existed for a while.

    .178, .179:
    
    Sounds like good advice.  I'll report back when we've tried your
    suggestions.
    
    Dick

    
    Remember also that the GF is the best thing since the light bulb. It's
    trying to tell you that you have a problem.
    

	Sorry if this was already asked in .25-.181, but does anyone
	know if such a thing exists as a GFCI outlet where you can wire
	such that one of the two plugs on it can be switched, while
	the other is unswitched?

	I could of sworn I saw on an older episode of "Home Time" where
	Joanne put in a GFCI outlet under the sink to plug in a
	dishwasher and garbage disposal, where of course the disposal
	would be switched, while the dishwasher is not.

	Home Depot sells no such unit that can be wired in this fashion,
	and the person I talked to had never heard of such a thing.

I've never seen such a thing.

		Steve

     I too tried to find this set up but no luck. It was for a bathroom
    sink area with outlet/light switch comination. The electrician
    suggested connecting to a GFI circuit breaker at the panel instead.
    	

    Well, you could wire a receptacle downstream from the GFCI, such
    that one plug of the downstream receptacle is switched and the 
    other isn't.  Or place the second receptacle next to the GFCI
    and use the GFCI for unswitched power and the adjacent receptacle
    for switched power.  You'd route the black wire out of the GFCI
    to the switch, and from their to the black connection on the
    receptacle.
    
    	Enjoy,
    	Larry