Replies

1. 
   pm22 | Dec 27, 2006 04:33am | #1

   The key word is "grandfather". You don't have to do anything if the house was wired to the Code that was in effect at the time it was built.

   As for the grounding, look around and notice which appliances have grounded, three wire plugs. Most likely computers and surge protectors. Also florescent lights need to be grounded. So concentrate your efforts there.

   For the sake of safety, you could add GFCIs to the bathroom and kitchen circuits. Being old, you don't have to worry about the silly AFCIs. As for nicked wires and that sort of thing, investigate and repair as necessary. GFCIs don't need a ground to function but may need it to test.

   ~Peter

1. 
   BillHartmann | Dec 27, 2006 05:50am | #2

   "GFCIs don't need a ground to function but may need it to test."No GFCI's don't need a ground for the test button to work.However, they do need a ground for the external testers to work.But the only approved way to verify the GFCI is with the internal test button..
   .
   Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.

1. 
   pm22 | Dec 27, 2006 07:52am | #5

   Thanks for clearing that up.

   Also the ground wire -- if you add one separately -- must eventually connect to the originating panel

   ~Peter

1. 
   BillHartmann | Dec 27, 2006 08:19am | #7

   "Also the ground wire -- if you add one separately -- must eventually connect to the originating panel"Almost. I believe the wording says that it has to connect to the grounding electrode system or something that connects to it.So I believe that you can connect to the grounding electrode wiring or the electrode, such as a cold water pipe within 5ft of where it enters the building.But most common would be to connect to the ground bus in the panel or the EGC in another box that in turn connects to the panel..
   .
   Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.

1. 
   cameraman | Dec 28, 2006 09:55pm | #10

   Bill, here in Michigan when we upgraded our house, not the entire house was up graded, so I ran a seperate ground  from some of the key recepital that I want grounded, back in to the new panel ground bar.  Our state inspector didn't want that wire in the panel. I ran it out to the ground rod and clamped it there, inspector was happy, end of story!

2. 
   DanH | Dec 27, 2006 06:31am | #3

   In most jurisdictions it's legit to pull a separate ground wire for existing outlets. Care must be taken with regard to the choice of grounding point, though. Grounding is needed for surge protectors and some electronics.

   A ground is unnecessary for the proper operation of a GFCI.

   People never lie so much as before an election, during a war, or after a hunt. --Otto von Bismarck

   
   

   ---

   Edited 12/26/2006 10:37 pm by DanH

3. 
   DanH | Dec 27, 2006 06:36am | #4

   Re bad connections, nicked wires, etc, if the wiring has been fine this long, and is not disturbed, there's no reason to believe that it's any more likely to fail than wire in a new home. (In fact, probably much less chance of failure.) The wire is too old to be the highly hazardous aluminum stuff (though be alert for any newer work with Al wire).

   The new arc-detecting breakers will protect against some wire-failure hazards, but not all, and are probably not justified unless there's some specific reason (you're not telling us) to be worried about the wiring.

   People never lie so much as before an election, during a war, or after a hunt. --Otto von Bismarck
4. 
   Omah | Dec 27, 2006 07:55am | #6

   Abandon it or rip it out, and put all new wires and panel in. No question.

5. 
   mikeingp | Dec 28, 2006 09:35pm | #8

   I've seen it here, frequently, that GFCI's don't need a ground to work properly, and I understand the theory. However, does that mean you are suggesting that standard GFCI's be installed when there is no ground? It seems to me that installing an outlet that pretends to have a grounding connector when it doesn't would not be a good idea.

   Also, does anyone know the rational for "the only approved way to test a GFCI is the internal test button"? I would prefer an external tester, since the internal tester seems a bit to me like putting the fox in charge of the hen house.

1. 
   BillHartmann | Dec 28, 2006 11:07pm | #11

   "However, does that mean you are suggesting that standard GFCI's be installed when there is no ground? It seems to me that installing an outlet that pretends to have a grounding connector when it doesn't would not be a good idea."ABSOLUTELY.In fact it is specifcally called out in the NEC as a way to replace ungrounded receptacels with grounding style.However, you are suppose to put a label on the GFCI and any downstream grounding style. IIRC it says "No Equipment Ground". They come with individual packed GFCI's."Also, does anyone know the rational for "the only approved way to test a GFCI is the internal test button"? I would prefer an external tester, since the internal tester seems a bit to me like putting the fox in charge of the hen house."The basic test is a resistor calibrated to 6 ma fault on the GFCI. It is connected to the to the hot lead on the load side to to the neutral on the line side. So it EXACTLY duplicated the type of fault that a person would find if they got tangled in a fault.I have been told, but not verified, that the testers can draw a much higher current.But with the newer GFCI's and specially those made after July 2006 the test button does more.It also is mechancially tied to latching relay that powers the receptacle. You can trip it with the test button even when power is not applied.I am not sure of the details, but part of it is part of a failsafe test that is finished with the reset button.Also it won't reset if the GFCI is installed "backwards" (with power connected to the LOAD terminals). That is one of the reason that GFCI's are shipped Tripped.
   .
   .
   A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.

1. 
   pgproject | Dec 29, 2006 12:34am | #12

   Our inspector passed a pair of kitchen outlets that were grounded by a separate wire to a clamp on a copper cold water pipe under the sink. I was surprised, as there would be nothing to stop a plumber from forgetting to reconnect the clamp in the future. Perhaps because these were GFCI outlets, the grounding issue became moot? This is a 1904 building with knob-and-tube wiring. All other circuits in the kitchen were new home runs to the panel. And yes, code states that a GFCI can be used on a two-wire circuit. I suppose the rationale is that the GFCI function will protect you regardless of the fact that there's no ground.

   Edited 12/28/2006 4:37 pm ET by PGproject

1. 
   BillHartmann | Dec 29, 2006 12:57am | #13

   "I was surprised, as there would be nothing to stop a plumber from forgetting to reconnect the clamp in the future. Perhaps because these were GFCI outlets, the grounding issue became moot?"There are a couple of other problems with doing that.First someone can get electrocuted when working on the plumbing system if there is current through the ground and the plumber cuts the pipe.The other is that the copper pipe might get replaced by plastic at some point. In both of those cases the work might be far from the ground connection under the sink and thus not be aware of it.That is why the ground electrode conductor or bonding conductor can only be attached within 5ft of where it enters the building.
   .
   .
   A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.

1. 
   DanH | Dec 29, 2006 01:34am | #14

   Howsomeever, I think in a lot of old building retrofit situations your more "reasonable" inspectors will be happy to see any improvement in immediate safety, regardless of the theoretical hazards down the line. In such situations they are generally allowed to use their discretion re how tightly to enforce the letter of the code vs the costs involved.
   Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot

2. 
   BillHartmann | Dec 29, 2006 03:57am | #15

   I agree in concept, but not in this case.I know that it used to be common to pickup a ground where ever one could line a piece of metal.I am not sure and don't have the old code books, but I am under the impression that this was a practice that was specficially removed from the code because of those kind of problems..
   .
   A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.

3. 
   pgproject | Dec 29, 2006 04:22am | #16

   "The other is that the copper pipe might get replaced by plastic at some point."Not too likely, as plastic is illlegal here. I guess it would be dangerous if the ground conductor was carrying current after the pipe it's strapped to is disconnnected from the (grounded) plumbing system, then the plumber grabs that pipe and a grounded pipe.Bill

6. 
   bobl | Dec 28, 2006 09:47pm | #9

   my house was built in 1960

   there is no apparent ground in a box. the ground wire was connected to the outside of the box, thereby grounding the box.

    

   bobl          Volo, non valeo

   Baloney detecter    WFR

   "But when you're a kibbutzer and have no responsibility to decide the facts and apply the law, you can reach any conclusion you want because it doesn't matter." SHG

7. 
   Allon | Dec 29, 2006 08:23am | #17

   Let's start with this: I am not an electrician.

   That being said, I believe that in the panel in my house the white is connected to the ground. Conversely couldn't he wire the ground in the existing outlets to white?

   Please weigh in because I'd like to learn too.

   -Allon

1. 
   BillHartmann | Dec 29, 2006 09:04am | #18

   "That being said, I believe that in the panel in my house the white is connected to the ground. Conversely couldn't he wire the ground in the existing outlets to white?"At ONE point and ONLY ONE POINT the neutral is bonded to the ground electrode system.That has to be someplace at the service entrance. It can be in the meter or in the main panel.The equipment grounding conductor is designed to never carry current in normal operation.That is one of the reasons that the neutral conductor can't be used as the ground.If you connect the ground on a receptacle to the neutral and if the neutral damage the ground connection can become hot. And likewise the metal case of any equipment that is plugged in to that receptacle..
   .
   A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.

1. 
   Allon | Dec 29, 2006 09:18am | #19

   Thanks for explaining. What I don't quite understand is this: If the neutral is connected to the ground at one point, doesn't that infact connect all of the neutrals to ground? Other than the natural resistance of the long path between the outlet and connection point, what is the difference?Thanks.-Allon

1. 
   DanH | Dec 29, 2006 03:26pm | #20

   You are correct from an electrical schematic standpoint, but not from a structural one. There are three basic concerns that are too difficult to explain briefly:1) (Most importantly) bad things can happen (using your scheme) if a neutral wire becomes disconnected between outlet and panel.2) Corrosion can be encouraged by multiple connections.3) "Ground loops" and other related effects can interfere with electronic devices.

   Half of the harm that is done in this world is due to people who want to feel important. They don't mean to do harm but the harm does not interest them. --T.S. Eliot