phil@diablo.amd.com (Phil Ngai) (08/24/89)
In article <5057@teklds.CAE.TEK.COM> brucem@teklds.CAE.TEK.COM (Bruce McAlary) writes: |I have done some re-work of older 2 wire systems, code allows you to wire the |neutral to the ground screw to on 3 wire plugs to satisify the ground. If you |examine a 3 wire system, the ground and white neutral are attached to a common |point in the breaker box, thus you are accomplishing the same thing by wiring |a jumper from the white neutral to the ground. Make sure you have the Didn't we just go through this? As I understand it, this is misinformation. As Henry Spencer likes to say "NEUTRAL IS NOT GROUND!" I believe that to do it right, you need a separate ground wire, of ampacity at least equal to the hot and neutral wires. Otherwise, a short from hot to ground could end up burning out the ground wire before the breaker blows. Tying the socket's ground and neutral together and only using a neutral wire exposes you to the voltage drop across the neutral wire which can be substantial. -- Phil Ngai, phil@diablo.amd.com {uunet,decwrl,ucbvax}!amdcad!phil "Today surgeons are highly respected but they were once just grave robbers."
smb@ulysses.homer.nj.att.com (Steven M. Bellovin) (08/24/89)
In article <26924@amdcad.AMD.COM>, phil@diablo.amd.com (Phil Ngai) writes: } In article <5057@teklds.CAE.TEK.COM> brucem@teklds.CAE.TEK.COM (Bruce McAlary) writes: } |I have done some re-work of older 2 wire systems, code allows you to wire the } |neutral to the ground screw to on 3 wire plugs to satisify the ground. If you } |examine a 3 wire system, the ground and white neutral are attached to a common } |point in the breaker box, thus you are accomplishing the same thing by wiring } |a jumper from the white neutral to the ground. Make sure you have the } } Didn't we just go through this? As I understand it, this is misinformation. } } As Henry Spencer likes to say "NEUTRAL IS NOT GROUND!" } } I believe that to do it right, you need a separate ground wire, of } ampacity at least equal to the hot and neutral wires. Otherwise, a } short from hot to ground could end up burning out the ground wire } before the breaker blows. Tying the socket's ground and neutral } together and only using a neutral wire exposes you to the voltage drop } across the neutral wire which can be substantial. As I recall the NEC, the ground wire does not have to be the same size; it won't be carrying the current for that long... However, you're correct for 12 and 14 gauge wires; it's only for the larger wires that the smaller size will do. On your main point, you're quite correct, as far as I know. However, many older houses, though equipped with 2-prong receptacles, actually have a ground conductor. If the house was wired with Romex -- and I've seen it in a house built ~1950 (it may have been as late as 1955, but I doubt it) -- there is quite likely a ground conductor inside the box. If the wiring is BX -- more likely in a house that old -- the metal sheathing of the BX is considered an adequate ground in the U.S. I believe that Canadian electrical code require a separate ground wire even with BX, but I've never done any wiring there... Either way, check it! Using a voltmeter, verify that there is ~120 volts between the hot (black) wire and the outlet box. If there isn't, that outlet is not groundable without running a new line to the breaker box. I've said a lot about GFCIs in the past; let me add one more comment here. The instructions I've seen with GFCIs have stated that they can be wired up to replace ungroundable outlets. That seems right in principle -- they're measuring current flow in the hot and neutral wires, not the potential to ground. But check -- your mileage may vary. The biggest problem in installing a GFCI is likely to be room -- GFCI outlets are quite bulky, and you may have trouble fitting them into a box, especially if there are other devices downstream. In that case, you may want to contemplate using GFCI circuit breakers instead. If you update the panel box, as someone else suggested might be needed, that can easily be done at the same time. Or it's easy enough to do yourself, though I confess that working inside the main breaker panel makes me a bit nervous. Other than the usual cautions -- and turning off the main breaker! -- figure out what brand to get; breakers must be compatible with the box (called a "load center"), and there are at least two or three different types. --Steve Bellovin
dya@unccvax.UUCP (York David Anthony @ WKTD, Wilmington, NC) (08/24/89)
In article <12073@ulysses.homer.nj.att.com>, smb@ulysses.homer.nj.att.com (Steven M. Bellovin) writes: > As I recall the NEC, the ground wire does not have to be the same size; > it won't be carrying the current for that long... However, you're > correct for 12 and 14 gauge wires; it's only for the larger wires > that the smaller size will do. If you value your life, you will make that ground conductor capable of carrying the full fault current so that the breaker trips first. > [there is an easy way to ] > check it! Using a voltmeter, verify that there is ~120 volts between > the hot (black) wire and the outlet box. If there isn't, that outlet > is not groundable without running a new line to the breaker box. No, no, no, no, no, no, no, no, no, NO! Verifying that there is 120 volts between the hot wire and the outlet box with an ordinary voltmeter (especially the very high impedance meters in the trade today) proves nothing more than the resistance of the ground path is not a significant fraction of the instrument's input impedance. Even for a 2000 ohm/volt instrument (which is very low impedance) a ground path resistance of 20 ohms would not be noticed as a voltage drop. Trusting your voltmeter to make a critical decision such as this is not, as my father would say, "erring on the side of prudence." There are two or three good ways to verify the ground circuit performance, but the general public should consult either an appropriate PE or licensed electrician. I've torn out enough armoured cable in radio stations to know that the ground path is not reliable. It is almost always high enough in resistance to completely screw up "single point bonding" in high RF fields (where the studios and transmitter are co-located). If you are in any doubt as to ground performance, consult an electrician. You are safest if you either install dedicated ground wiring, or replace the branch circuit in question in its entirety. The latter is what we are doing in our home after my wife's encounter with an alleged "armoured cable" ground path and a very much live refrigerator frame. York David Anthony WKTD Wilmington, NC P.S. I didn't know that the cable in some parts of our house was armoured and some was Romex (it had two additions made since original construction). If you are buying an older home, two or three hours of an electrician's time to assess the condition of the electrical plant is money very well spent. Rewiring costs $$$$$$$$$$$$$$.
hgw@julia.math.ucla.edu (Harold Wong) (08/24/89)
In article <26924@amdcad.AMD.COM> phil@diablo.AMD.COM (Phil Ngai) writes: > >As Henry Spencer likes to say "NEUTRAL IS NOT GROUND!" > >I believe that to do it right, you need a separate ground wire, of >ampacity at least equal to the hot and neutral wires. Otherwise, a >short from hot to ground could end up burning out the ground wire >before the breaker blows. Tying the socket's ground and neutral >together and only using a neutral wire exposes you to the voltage drop >across the neutral wire which can be substantial. > I was once told that I could use the metal conduit as the ground and eliminate the third wire. Is this true? Is this safe. I have flexible aluminum conduits running to all of my boxes in the house. ------------------------------------------------------------------------------- Harold Wong (213) 825-9040 UCLA-Mathnet; 3915F MSA; 405 Hilgard Ave.; Los Angeles, CA 90024-1555 ARPA: hgw@math.ucla.edu BITNET: hgw%math.ucla.edu@INTERBIT
alanj@nevermore.WV.TEK.COM (Alan Jeddeloh;685-2991;61-201;292-9740;orca) (08/25/89)
It is not necessary to run the grounding conductor all the way back to the
load center. The NEC allows that, when doing "old work" (modifications
to an existing installation), the box may be grounded by running a separate
grounding conductor to the nearest good ground. This can be another box
that is properly grounded, a metal *cold* water pipe, metal building frame,
conductor buried in the foundation, separate ground rod, etc., etc., etc.
Waste (sewer) pipes, gas pipes, plastic(!) water pipes are not considered good
grounds (the NEC does allow gas pipes, but most local codes forbid them.
In this area (Portland, OR) new gas installations use plastic gas pipes to
run to the house, so the gas pip in the hose is not a good ground, anyway.)
The new grounding conductor must not be less than #12 and at least some
(forget which) percentage of the size of the hot/neutral pair. The grounding
conductor must be properly protected from damage.
IT IS NOT LEGAL OR SAFE TO CONNECT THE NEUTRAL TO THE GROUND!!!!
-Alan Jeddeloh (503) 685-2991
Tektronix ITD Networking; D/S 61-201; PO Box 1000; Wilsonville, OR 97070
{decvax|ucbvax}!tektronix!orca!alanj -or- alanj@orca.wv.tek.com
This space available for Bush's first blunder.marulli@kodak.UUCP (Dave Marulli) (08/25/89)
In article <12073@ulysses.homer.nj.att.com> smb@ulysses.homer.nj.att.com (Steven M. Bellovin) writes: > > >........................................................... However, >many older houses, though equipped with 2-prong receptacles, actually >have a ground conductor. If the house was wired with Romex -- and I've >seen it in a house built ~1950 (it may have been as late as 1955, but >I doubt it) -- there is quite likely a ground conductor inside the box. My house was built in 1956 (so was I!) and the above is true and untrue. The first floor had Romex witha ground to the box, which I easily attached to new, grounded outlets. The second floor has Romex with no ground and I would have to pull a new wire to add a ground upstairs. >I've said a lot about GFCIs in the past; let me add one more comment >here. (...some stuff deleted...) >..... The biggest problem in installing a GFCI is likely to be room -- I have no verification of this, but the guy behind the counter at my local electrical supply house mentioned another problem with GFCI outlets: Let's assume the bathroom outlet that you want to replace with a GFCI is the first outlet on the circuit. He said that depending on how many other outlets are on that circuit, you may get what he called "annoyance trips". i.e. bedroom A.C. units, refridgerators, etc may cause the GFCI to trip, when a regular circuit breaker would not. I'm not sure of the reason behind this, and I know how a GFCI works. I'm just passing this along.
jordan@Morgan.COM (Jordan Hayes) (09/01/89)
Harold Wong <hgw@math.ucla.edu> asks: I was once told that I could use the metal conduit as the ground and eliminate the third wire. Is this true? Let me relate a little story. A few years back, I had 3-prong outlets that (as I later found out) were connected to a 2-wire plant, with the ground attached to the conduit. Everything was fine for a while. I had a stereo, VCR, tape deck, TV, etc. plugged into this outlet -- all plugs had 2-prongs. One day I got a new TV with 3 prongs, and plugged it into the loop. It turns out that somewhere along the line, the hot wire had a short to the conduit. Now, I also had cable TV, which *knows* what ground is (I could see it plugged into the dirt outside my window!). A fire-fight ensued between the Cable TV and the VCR. "Jordan, do you think that thing should be smoking and sparking?" -- my old roomate, leslie The VCR lost :-( /jordan
wyatt@cfa.HARVARD.EDU (Bill Wyatt) (09/02/89)
> I was once told that I could use the metal conduit as the > ground and eliminate the third wire. Is this true? > [...] A few years back, I had 3-prong outlets > that (as I later found out) were connected to a 2-wire plant, with the > ground attached to the conduit. Everything was fine for a while. [...] > It turns out that somewhere along the line, the hot > wire had a short to the conduit. [...] Bad news, but this might have happened with separate ground cable as well, since the ground wire is usually bare. With hindsight, the proper solution is to buy one of those simple testers with LEDs that detect this sort of short, and to use it periodically on each of your outlets. Bill Wyatt, Smithsonian Astrophysical Observatory (Cambridge, MA, USA) UUCP : {husc6,cmcl2,mit-eddie}!harvard!cfa!wyatt ARPA: wyatt@cfa.harvard.edu SPAN: cfa::wyatt BITNET: wyatt@cfa