Electrical Ground Problem

[cabinetman]

We ordered U-verse, and one of the requirements is that the outlets must be grounded. One of the outlets tested "open ground". The outlet is screwed to a metal box, inside an interior block wall. I'm pretty sure there's conduit.

I added a ground wire to the outlet and screwed it to the box, still tests "open ground". Any suggestions?
.

[LCHIEN]

if its not too old, most electrical house wiring will have 12-2 or 14-2 romex with a copper ground wire, and no conduit. The green ground wire should be connected to the green ground screw on the outlet. Is there no ground wire?

[cabinetman]

if its not too old, most electrical house wiring will have 12-2 or 14-2 romex with a copper ground wire, and no conduit. The green ground wire should be connected to the green ground screw on the outlet. Is there no ground wire?

It's a 2 wire 12 ga romex, no ground wire.
.

[Black wallnut]

why are you "pretty sure" there is conduit? I'm betting there is not which means you'll have to pull a ground wire to the box.

[Tom Slick]

How old is the house? My house (built in the '60's) has cloth covered 14-2 romex with the ground wire wrapped around one of the nails that hold the box to the stud. You don't see the ground in the box.

With a multimeter you can put one lead to neutral and the other to the box, if you have continuity then it is grounded. Otherwise you'll need to run a separate ground. You should also double check all of the connections at the neutral/ground buss bar in the panel to make sure they are tight and not corroded.

[ivwshane]

As a side note: I wish I could get u-verse in my area

[cgallery]

Pulling a ground wire through an interior block wall doesn't sound like much fun. Do you have access to the top of the blocks (like through an attic space)?

[eccentrictinkerer]

A GFCI outlet will satisfy the NEC code (and most local codes) for a 3-wire outlet (even without a ground).

Obviously, you still won't have a physical ground. Usually cable and satellite installers install a redundant ground. Around here anyway.

[cabinetman]

Pulling a ground wire through an interior block wall doesn't sound like much fun. Do you have access to the top of the blocks (like through an attic space)?

No I sure don't. I may have an out though if push comes to shove. The outlet is behind a cabinet that sits against the back wall and an outside wall. There is the main water line to the house about 20ft from the location. I could hook a ground wire to the piping and surface mount it to the inside next to the outlet. Then, poke a hole through the wall next to the outlet, and connect it to the ground (green) screw. How does that sound?
.

[Tom Slick]

You could do it, and it'll work, but it's not allowed by the NEC anymore.

[woodturner]

With a multimeter you can put one lead to neutral and the other to the box, if you have continuity then it is grounded.

That is not a valid test for ground. All that shows is DC continuity between the neutral pin and the ground pin. A minimal DC connection that will not withstand the amperage of a ground fault will show as connected with a meter. In the event of a ground fault, however, that "connection" may well be an open circuit and kill someone.

As another poster mentioned, a GFCI can be used to provide a virtual ground to protect life, and is compliant with NEC. However, it will not protect electronic equipment adequately.

Sounds like the OP would be better off to hire a licensed electrician.

[Tom Slick]

It would so if some form of a ground exists vs. no ground at all, which is what he is trying to figure out.

I agree he needs an electrician.

[woodturner]

It would so if some form of a ground exists vs. no ground at all, which is what he is trying to figure out.

I agree that the meter will show a connection if a low current DC connection exists to ground. What it does not indicate is whether that connection is a relatively poor ground, for example through the sheathing of BX (armored cable) or whether it is a good ground, for example the ground wire in NM-B cable ("Romex").

When there is an actual ground fault, tens or hundreds of amps may travel through the ground path. If the ground is a poor ground, such as the screw on the cable clamp holding the BX cable, the connection acts like a fuse and blows, leaving no ground path. In that case, the result can be hazardous to people. If it's a good ground, though, the breaker pops and everything is fine.

The problem with using the meter is that it cannot tell you if the connection is a good one or a bad one. Same issue with the plug in receptacle testers, BTW - they will indicate the presence of a ground, even a poor ground.

[Uncle Cracker]

A good ground must be able to handle significant current, or it won't do its job. It might have to handle a shunted surge discharge from a lightning protection device, or appear as a more attractive path for a shorted tool than your body does. A multimeter does not generate enough test current to determine if the ground path is substantial. It takes specially made equipment to do this. Otherwise, your life, and that of your equipment, may be resting on a hair-thin scrap of stranded wire, a metal shaving, or even a staple (seen it myself). Be safe, be sure.

[phi1l]

This is something you will want to get done right, not just for proctection, but also to minimize ground loop hum from your audio equipment.