Re: Replacing 2-Prong Receptacles

eg wrote:
On Jul 10, 9:54 am, bud-- <remove.BudN...@xxxxxxx> wrote:
eg wrote:
On Jul 8, 11:23 pm, bud-- <remove.BudN...@xxxxxxx> wrote:
eg wrote:
On Jun 5, 8:16 am, Bud-- <remove.BudN...@xxxxxxx> wrote:
RichardandPeg wrote:
On Jun 4, 4:28 pm, Dr J R Stockton <j...@xxxxxxxxxxxxxxxxxx> wrote:
In sci.engr.electrical.compliance message <1180923644.541838.311450@p47g
2000hsd.googlegroups.com>, Sun, 3 Jun 2007 19:20:44, RichardandPeg
<beezob...@xxxxxxxxxxx> posted:
I hope this isn't too dumb...
I am replacing some old 2 prong receptacles with new 3 prong ones.
Still only the old two conductors-hot and neutral- in the box. At
least I don't have to run all over the house searching for an adaptor
when I want to plug something in.
Actually I am a pretty good electrician even tho that isn't my job.
My assesment of the deal was that a three prong receptacle in a two
wire box is no different than a two prong receptacle with a plug
adaptor. When all is said and done what I have is still only two
wires- hot and neutral- powering what ever is plugged in.
A GFCIoutlet (or downstream from aGFCI) makes it safe. If the current
in the 2 wires differs by 5mA (like current flowing through your body to
ground) theGFCIwill trip.
Can someone please explain how aGFCIoutlet, where no ground is
present, works? I've seen this described many places (besides here);
however with my (albeit limited) knowledge of what "grounded" and
"GFCI" mean, I don't undertand how this works (I understand that the
GFCIoutlet would have to be wired in a particular way, but I still
don't "get it").
All GFCIs compare the current in the "hot" and "neutral" wire. If there
is a difference of 5mA theGFCItrips. If the currents are not the same,
some of the current is flowing through an alternate path, such as
through you to ground (as explained above). The comparison of the
currents is done by running both wires through a "current transformer".
The ground wire is not needed and not used in the detection. TheGFCIis
wired just like a regular outlet (except on 2 wire circuits there is no
ground path available).
Also, it seems like using aGFCIoutlet, where no
ground is available, is only a "backup solution", and that an actual
ground is the ideal solution. If so, in what ways are one better than
the other? In particular, which one protects the equipement, the
user, the house from burning down, ...
It is certainly a good idea to have a ground connection at the outlet. A
ground connection keeps the potential of exposed metal at approximate
earth potential. But another major function of a ground wire is to
provide a low resistance path back to the power source to trip a breaker
when a hot wire faults to the exposed metal. (In most power systems the
neutral is connected to the ground wires.) This provides shock and fire
protection.
GFCIs provide additional protection in locations where contact to metal
connected to earth, like plumbing is readily available. Or where where
earth is available as outside or through concrete floors. This may
provide fire protection but is primarily aimed at shock protection.
On 2 wire circuits where there is no ground path back to the panel, it
would typically be quite expensive to add a ground path. But many
devices have a 3-pronggroundedplug. The US-NEC allows a 'grounded'
typeGFCIoutlet to be installed, which gives significant protection
using agroundeddevice on the circuit without the ground path.
Also, my understanding is that
a non-groundedGFCIhas to have special wording on it (by code). What
is this wording? Any outlet that has to come with a disclaimer sounds
scary to me...
For aGFCIinstalled on a 2 wire circuit (no circuit ground available)
install the sticker supplied with the outlet "No equipment ground". That
warns a user that thegroundedtype outlet does not actually have a
ground connection. Foroutletsinstalled downstream from aGFCI
(protected by theGFCI) also add the label "GFCIprotected".
Further questions?
Sure - if you're willing to answer them. (The context - my 5 year old
son's room has 3 "grounded"outletsthat indicate "open ground", when
I plug in a circuit tester. I have a n electrician coming out soon.
He indicated that ground might be available in theoutletsor that he
might be able to do this "non-grounded-GFCI" trick. It might be
possible, albeit expnsive, to run "groundedconduit" (not sure if this
is the best term) along the outside of the house from a pre-existing
ground source. If this sounds strange, it's done in Eichler style
homes in California all the time).
If your ground tester indicates "open ground" there *probably* is no
ground available at the outlet, andgrounded-typeoutletsshould not
have been installed. I have never seen grounding added on the outside of
a building. If theoutletsare on the 1st floor and there is a basement
with an open ceiling, ground wires may not be too hard to add. Else
adding ground wires is a major pain. If grounds are added, they *must*
tie to the rest of the system, not an isolated ground rod.



So I'm still trying to understand what incremental benefit a real
groundedoutlet provides over a non-groundedGFCIoutlet. If a short
occurs between hot and the metal casing of a computer, does the
groundedoutlet provide better protection for the computer?
In my opinion depends on the nature of the short - probably not. In a
metal case computer, the problem short would only occur in the power
supply and is not likely.

Better
protection for the person touching the metal casing?
In general yes. With a solid short the circuit breaker will trip. But aGFCIwill provide shock protection. To get a shock you need not only the
computer case but a path to 'ground'. In your son's room that might not
exist (no path, no shock - like birds sitting on a power line). The NEC
allows GFCIs to be used because they provide reasonable protection where
no ground wire is available. Life is a series of trade-offs.

Better protection
for the circuitry in the house?
The circuit does not know there is a short until there is a path from
case to 'ground'. If there is a path, theGFCIwill trip.

You suggest that in a non-grounded
GFCIoutlet, there will not be that "low-resitance path back to the
power source". To me that suggests that if a short occurs, there will
be a lot of current that has nohwere to go (before theGFCItrips)
except through something with high resistance causing too much wattage
for the wiring and leading to something burning...
There may be voltage 'with nowhere to go'. But if there is high
resistance there will be no current. You could have an electrically
caused fire on agroundedor un-groundedcircuit. Probably less likely
with aGFCIthan just agroundedcircuit.

Other possibilities:
Just use devices with non-groundedplugs in that room.
I'm not sure if laptops have a connection from the power ground to the
power circuit to the laptop. Or laptop with ungrounded plug.
Locate computers with ground plugs elsewhere - assuming there areoutletswith a real ground.

If I had a kid in a room with un-groundedoutletsI would probably trust
aGFCI, particularly if there were no ground paths in the room to get a
shock. I would probably use wireless phones and wireless routers, but
that is probably overly protective.

Note that GFCIs have a test button and should be tested about once a
month and probably after powerline 'events'.

Further questions?

--
bud--- Hide quoted text -

- Show quoted text -

First - thanks for taking the time to answer. More followups if you
have the time.

Several times you predicated you responses with "if there is a path to
ground". My understanding is that if a person touches something that
has shorted out to "hot" (assuming no grunded outlet), then there is
always a path to ground - that path being from "hot" through the
person's body, through the person's feet. And being that the human
body doesn't have that much resistance, enough current will flow
though ones body to do some real harm. And while the amount of current
might depend on the surface of the floor (carpeted, ...), type of
shoes, ... there's always "some" path to ground. Is my understanding
here incorrect?

Wood floor, carpet, floor tile - negligible current. All are good insulators. Touching the insulation on a 'hot' wire does not give you a shock.

Grounded plumbing as in a kitchen - problem. Concrete basement floor - problem.



Alternatively, here is my understanding of what happens when something
shorts and is:
1) plugged into a grounded ooutlet
2) plugged into a non-grounded GFCI
Please let me know if I understand this correctly.
In (1), you've suddenly created a path from "hot" to ground (i.e. from
the location of the short through the 3rd prong) with very little
resitance. So a lot of current starts flowing, immediately (one
hopes) tripping the circuit in the main panel. Once that happens, the
dangerours part is past. Ideally this happens when one isn't using
the electrical item (so there is no shock), but even if someone is
using the electrical item, the circuit trips quickly enough to avoid
bodily harm.

Yup

In (2) the metal that has been shorted to "hot" suddenly goes from 0
volts to 110 volts (and can stay that way as long as no one touches
the electrical item). Then as soon as someone touches the exposed
metal that is now "hot" current starts flowing through that person's
body; however, the GFCI outlet detects that hot and neutral have
different amounts of current and it trips, shutting off power to the
outlet. Hopefully, the GFCI outlet trips fast enough that the amount
of current that went through your body didn't do any damage. But in
any case, you are likely to feel at least some shock.

True if there is a path to ground. If there is no ground path for the person, there is no shock (and no GFCI trip). As noted above, there is no ground path in most of a house. In a 2nd floor bedroom you probably have no ground path, and touching a 'hot' case won't do anything (like birds sitting on a power line).


Based on you last comment "I would just a GFCI if there were no path
to gorund... would probably use wireless phone and wireless
router...". Again, I though that there was always a path to ground -
and I'm not sure why wireless would be safer than wired... so I'm
guessing that my understanding is not totally correct...

Often no path through person as above. Wired phone is a possible path (but not likely).

--
bud--
.