Bathroom GFCI with no ground?

Page 1 of 2  

I want to replace an existing 2-prong receptacle in a bathroom with a GFCI receptacle. There is no ground wire going to the existing receptacle and no grounded metal cable or grounded metal outlet box -- just two separate wires, one white and one black. There is also no practical way to run a new ground wire back to the service.
In this situation, is it okay to go ahead and install the GFCI receptacle and put the sticker on it that says "No Equipment Ground"?
UP.SO. I "could" run a ground wire from the GFCI receptacle to a cold water pipe and make sure that there is a jumper across the hot water tank in and out pipes and across the water meter. But, I think I read somewhere about that being a bad idea (using the cold water pipe for a ground). That is a bad idea, right?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Jay-T wrote:

Yes
Used to be code compliant up to about 10 years ago. It isn't now since existing metal in the path back to the system ground can be replaced by plastic in the future.
--
bud--

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Use GFCI because it operates on the 'difference' in current between live and neutral. If someone did get a contact or drop the hair dryer in the wet metal bath tub any conseuent current likely to trip the GFCI safely.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

agreed, but will the TEST button still work without a ground?
Mark
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Yes, as I understand it (I haven't tried it). I believe the TEST button in the receptacle allows a small current to pass from the LINE side hot to the LOAD side hot around the current transformer, thereby creating the imbalance that should trip the GFCI. In contrast a plug-in tester will not work, as it relies on the ground connection.
Cheers, Wayne
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 3 Feb 2010 20:35:57 +0000 (UTC), Wayne Whitney wrote:

And without a ground the GFI may not work properly.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Please explain.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
In wrote:

A GFCI will indeed work properly and to specs without a ground. Go read how they work and why/when they trip. Every box and most sites tell you what to do when there is no third wire. The OP had it right. Earth is NOT used in the operation of the GFCI.
http://www.doityourself.com/stry/gfci http://wiki.answers.com/Q/How_does_the_ground_or_earth_wire_of_an_appliance_protect_a_user_if_there_is_a_fault_with_that_particular_appliance or if you're Canadian http://en.wikipedia.org/wiki/Residual-current_device
Google is full of good information
HTH,
Twayne
HTH,
Twayne
--
--
Newsgroups are great places to get assistance.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Michael Dobony wrote:

Absolute nonsense, as GFIs measure only the imbalance of currents in the hot and neutral lines that pass through their current sensing transformers. No GFI can require an earth ground for proper operation and still meet UL standards.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Wayne Whitney wrote:

The test button merely connects the hot to the GFCI's ground reference, so no earth ground is needed to test operation. But a plug- in GFCI tester will work, whether or not the GFCI is earthed.
Here's National's document for their old LM1851 GFCI chip:
www.national.com/ds/LM/LM1851.pdf
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
do_not_spam snipped-for-privacy@my-deja.com wrote:

Right except it is line side hot to load side neutral (or vice versa).

Nope. The GFCI outlet circuit makes no connection to "ground". The test button connects a resistor from the hot wire ahead of the current transformer to the neutral downstream from the current transformer (or vice versa) resulting in a current through the transformer only on the neutral wire.

A plug-in tester connects a resistor from hot to ground. Since ground is not connected to anything a plug-in tester will not work.

Interesting that the circuit (fig 2) does not show the test button that is on all GFCIs.
Also interesting that the feature that causes an immediate trip if there is a downstream N-G connection only adds 2 parts.
--
bud--



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
bud-- wrote:

But I've tried it with a Hubbel (Shock Shield 3-outlet) plug-in GFI with its ground prong retracted and a Sears branded Leviton wall outlet GFI connected to the AC through a 2-wire cheater cord. I believe both used General Instruments GFI chips.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Yes.
The test button does not send any current to ground so no ground is required for it to work.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
DerbyDad03 wrote:

Thanks. I wondered about that too.
Looks like I won't have a problem going ahead with the GFCI even though there is no ground.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

BTW...not only is it OK to (and you should) replace the ungrounded bathroom receptacle with a GFCI, it is also code compliant to replace any 2 prong, ungrounded receptacle with a 3 pronged receptacle as long as that receptacle is protected by a GFCI and marked as having No Equipment Ground.
For example, if you have a circuit full of daisy chained 2 pronged, ungrounded receptacles, replace the first one in the circuit with a GFCI and wire all the downstream 3 prong, yet still ungrounded receptacles off of the load side.
It's a fairly common practice and allows for the use of three pronged cords in "older" homes.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
In wrote:

NO, it does NOT "allow" use of 3-wire equipment!! If a 3-wire piece of equipment is plugged into an outlet, the earth ground in that case DOES need to be present. THAT is why the outlets must be labeled as having NO ground! It's to prevent just that kind of accident waiting to happen.
HTH,
Twayne
--
--
Newsgroups are great places to get assistance.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Perhaps we need some clarification here.
We are talking about people protection only. We're not discussing whether the absence of a ground will be a detriment to the operation of the equipment.
It is my understanding that the GFCI will protect the user should there be current leakage to ground within the device.
If there is any type of short to ground within the device then there will be a differential between the current on the hot and the current on neutral and the GFCI will trip and the user will be protected.
Why else would the NEC allow for the use of a 3 prong outlet on an ungrounded circuit if they didn't expect you plug a 3 prong cord into it?
You'll note that code says it must be labeled as No Equipment Ground, not simply No Ground.
For the safety of us all, I won't mind if you correct me.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
DerbyDad03 wrote:

Downstream outlets are also tagged "GFCI protected".
Avoid connecting the outlet grounds together on downstream grounded outlets where a ground is not present.

Seems like that should have been obvious to Twayne. If 3 wire plugs were not allowed the NEC would require a 2 wire GFCI outlet - which it doesn't. The NEC explicitly allows 'grounded outlet' GFCIs on 2 wire circuits, and as you said above, allows the GFCI to protect conventional grounded receptacles downstream.
--
bud--

>
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Well, 406.3(B) of the 2008 NEC states: "Receptacles and cord connectors that have equipment grounding conductor contacts shall have those contacts connected to an equipment grounding conductor". And 406.3(D), the section that deals with using GFCIs on circuits without equipment ground, does not appear to override 406.3(B). So I agree the situation is confusing and the intent is not clear.
Cheers, Wayne
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Set Lawyer-Speak Mode = True
A 3 prong plug is typically connected to a cord which has an equipment grounding conductor within, therefore the "cord connector" portion of 406.3(B) is satisfied.
Once this cord is plugged into a 3 prong receptacle, the receptacle's equipment grounding conductor contact will be connected to the cord's equipment grounding conductor, this satisfiying the receptacle portion of 406.3(B). Prior to the insertion of the plug, the grounding hole in the receptacle is nothing more than an extra hole since it's not attached to anything. If it's not attached to an equipment grounding conductor, it isn't a equipment grounding conductor contact because the code says it isn't.
Set Lawyer-Speak Mode = False
That clears it up, now, doesn't it?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Related Threads

    HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.