Sign in with
Sign up | Sign in
Your question

Poor or No Electrical Ground - What to Do About It

Last response: in Components
Share
a b ) Power supply
July 27, 2009 6:31:28 PM

So you find yourself in a place (maybe a new apartment?) with apparently no Ground in the electrical outlets. This is probably because they are all 2-prong outlets, or maybe you've used an outlet checker and it says the Ground is no good. What can you do? By far the best solution technically is to re-wire the building, but let's look for the less expensive, more practical things if you are not the owner planning major home upgrades.

You have several issues to deal with here, starting from the apparent situation that there is no reliable Ground anywhere. A Ground is used for two purposes: part of a protection system for wiring or circuit faults that mistakenly send real power to the computer's chassis that is accessible to people; and removing static charges and electrical signal noise while they are at low levels, before they build up to troublesome.

1. Assume you have no Ground anywhere. Even if there are a few three-prong outlets in walls, they MAY have been especially installed properly, but quite possibly not. It's an old and BAD habit to replace a two-prong outlet on an ungrounded system with a three-prong one just to allow using that kind of plug, but NOT actually provide any Ground connection! So unless you can verify that the 3-prong one is done correctly, assume it is not. (See item 5 below for tests you can do with a volt / ohm meter.)

2. Here is how 110 VAC house wiring in Canada and USA (and others) is supposed to be set up. It's based on what is called a Grounded Neutral system. Both at the transformer out on the power pole, and at the fuse or breaker box in the house, the Neutral line is connected to a true Ground to the earth. So at those two points the Neutral line is at zero volts potential compared to the earth. Anywhere else in the house the Neutral line is NOT guaranteed to be the same as ground. It is carrying current supplied to the user's device (a computer or a tea kettle) back to the Neutral bus in the fuse box. Since the wiring in the wall has a very small but non-zero resistance, there actually is a small voltage present at the wall outlet on the Neutral line, compared to true Ground.

3. At the house fuse box or breaker panel there is a true Ground connection to the panel box itself. As I said, that also is connected to the Neutral bus in the box. In modern designs there is a separate Ground bus in the box, and all wiring running out to circuits everywhere in the house is done with cable containing two current-carrying conductors (insulated) plus one bare copper wire for Ground. This is how Ground is provided at every outlet box in the house. And in all of these, NOTHING (except a Ground lead) should be connected to that Ground, so it never carries any current under normal circumstances. In older systems there was no Ground lead extending beyond the main fuse box, and no Ground bus in the box. In a few older systems, though you may find a circuit has been added on later using what is called "BX cable". This is the wiring with a spiral metal covering on the outside. Normally the clamps used to anchor the cable to the fuse box and to the outlet box in the room manage to establish an electrical connection between the fuse box itself and the outlet box. In that case, because the fuse box itself is supposed to be grounded, the outlet box will be, too.

4. Today's design with 3-prong outlets etc. has a cable coming into a box in the wall with three wires. (The wire is labeled something like 14/2, indicating that it has 14-gauge wire in it and has two current-carrying insulated wires, PLUS a bare Ground wire). The color code is: Black for Hot (the "supply" of power), White for Neutral (for returning current) and bare for Ground. When a 3-prong outlet is mounted in the wall box so that, of the triangle of slots and a hole, the round hole is at the bottom pointing down, then the round (bottom) one is Ground, the left one with the wider slot is Neutral, and the right one is Hot. That way you cannot plug even a 2-prong proper "Polarized" plug in wrong, because the wider (Neutral) blade won't fit into the narrow (Hot) slot. In older non-grounded systems, there is no Ground, but there still are Hot and Neutral. However, often these sockets were made with no difference in slot widths, and there was no standardization on whether Hot was left or right.

5. In a modern system if you use a neon tester you will find 110 VAC from Hot (right) to Neutral (left), and also 110 VAC from Hot to Ground, but no apparent voltage from Neutral to Ground. That is because the tester cannot show you very small voltages between Neutral and Ground. In an older ungrounded system, you'll still detect 110 VAC from Hot to Neutral, but there appears to be no voltage from either of those slots to the metal box in the wall. That is because the metal box is not connected to anything! There is no Ground available at the box. If you have a decent multi-use electrical meter with AC Volts and Resistance ranges, you can do three other tests. The first is simply to repeat those neon tester ones, setting the meter range to measure more than 120 VAC (say, 0-200). Note the readings - they should look like either 110 to 120 VAC, or no volts. Now the second tests: for only the ones that appear to be zero volts (like Left (Neutral) to Ground), try setting the voltmeter to even lower ranges, like 0-5 VAC or 0-1 VAC. IF you also have a heavier load connected to the circuit (say, a teakettle plugged into the other half of the outlet) you may see small fluctuating voltages there. However, if you are testing an old system with no Ground in the box, you probably will still see no voltage just because the box is NOT connected to Ground or anything else. Now the third test is for Continuity or Resistance. You ONLY do this on places where your previous test says there is no measurable voltage present, so maybe the two test points are actually connected together somewhere. You set the meter to read Ohms Resistance and apply the probes to the two test points. If they are connected, you should read only a very few ohms, and you may find the reading depends on how hard you push the probes to make contact with the metal. If you think the place you are trying to test really is a Ground and want to verify that (for example, the metal box mounted in the wall, or the Ground (round) hole in an outlet), you need a known true Ground for reference. Try hooking up a long piece of wire to a water pipe or, even better, to the real Ground terminal where your fuse box is connected to a water pipe near it. Now measure the resistance from your test point to true Ground. It really should be no more than a few ohms, often less than one ohm and hard to measure.

6. Now to start on solutions, other than moving or persuading the landlord to re-wire the place. For the purposes of getting rid of low-level voltages, either static charges or electrical noise signals, you can try to connect a separate (green) wire from the computer outside case to a reliable Ground. Best place to look is a water pipe. Water supply systems run a pipe though a lot of soil, thus providing good electrical contact to the earth. Older systems are ALL metal piping, so it works. However, some have been modified with plastic piping components in odd places, so it is always possible that a water tap may NOT be a reliable Ground, and it is nearly impossible for an amateur to tell. But that's still your best bet. (See item 5 above on how to test and verify whether a possible Ground point near your computer really is Grounded to the earth.)

7. The other function of a good Ground is in electrical supply system safety. One way a Grounded system protects you is that any part of your appliance that is exposed to people should be connected to true Ground. Then if anything goes wrong in the appliance and the Hot supply comes into contact with that exterior, the Ground lead does two things. The most important is that it provides a high-current-capacity route for the power to be taken to Ground - so high that the current will exceed the limit of the fuse or Breaker in the panel that supplies the Hot lead, and it will blow out. That stops the current. The second is that, during the very short time all that takes, the exterior case you can touch will be VERY close to zero volts anyway, so you are unlikely to be shocked. Now, if you install a green grounding wire to a nearby water tap, it may or may not be able to carry all that current and provide the protection a real Ground system can - it depends on how good the substitute ground connection is. But it should be able to handle the small currents from electrical noise and static charges IF it is actually connected to the earth.

8. An alternative to a true Ground system for safety is a Ground Fault Circuit Interrupter (GFCI). You can actually buy the simpler one-circuit version of these at any electrical supplies or hardware shop. They look a lot like a normal 3-prong double outlet (but more expensive) and you mount them into the wall box in place of the old outlet device. (Obviously, before starting you MUST isolate power from the box before opening, by removing the fuse or shutting off the breaker that supplies this outlet box.) They are different on the front in that they also have a little indicator light, a TEST button, and a RESET button. If you don't understand house wiring you are best advised to have this item installed by someone who does, like an electrician. (If you are competent at home electrical systems, see end of this paragraph.) Once properly installed, what they do is to constantly measure and compare the currents flowing in both the Hot and Neutral parts of the circuit. Those ALWAYS should match. If they don't the GFCI interprets that to mean some current is leaking out of the circuit to somewhere else, and that could be a big problem. So it shuts off the power or "trips out", just like a circuit breaker does. Once you fix the original problem, you use the Reset button. You can test at any time with the Test button, and it will trip out for you. The intriguing part is that this device does not need a true Ground point to operate, so it can work in an ungrounded outlet box. It just uses a different measurement to decide when to shut off power. So it provides very similar safety protection for people and for many appliance malfunctions, but it does nothing for getting rid of electrical noise and static charges. Now, for those competent to install a GFCI, here are notes to use; all others skip down. READ the GFCI unit's manual; it usually has four terminals or wire leads. Three of these are the usual Hot (Black), Neutral (White) and Ground (Green or bare copper). These are the supply connections to the double outlet. The fourth (often Red) is an OUTPUT of sorts. It is a Hot connection on the Load side of the device (just like the small right-hand slots) that you can use to power additional outlet boxes further down the circuit layout. If there's a cable leaving this box to power other boxes, use this "Output" Red as the Hot supply going out on that cable. Then ALL the outlet boxes "downstream" from this box also are protected by the GFCI, but NOT any boxes that came before it. For that reason, if you plan on using the GFCI for this multi-outlet box protection system, the device should be mounted in the very first box in the chain - the one fed directly from the Fuse Box or Breaker Panel.

So, a combination of a GFCI for safety protection against malfunctions, and a ground wire from the computer's case to a water line for removing static charges and electrical noise, may work for you. These still do nothing for power system surges, but a modern grounded system does not, either. That's a whole separate issue.

More about : poor electrical ground

July 27, 2009 7:07:52 PM

Very nice. I learned all this a long time ago, and actually was wiring before ground faults existed. Hope this helps a few out there. Nice post
a b ) Power supply
July 28, 2009 12:11:05 AM

Yes indeed.
Related resources
August 4, 2009 12:19:12 PM

Just some points I'd like to clarify:

2: Yes there is/can be a small voltage apparent on the ground, but its so small it can be considered to zero volts.

4: The direction of current flow is arbitrary but its usually referenced in the positive direction which is from black (hot - aka postive) to white. I mean if you wanted to you could say that the current flows from white to black but is negative.

Great write up though. Very informative and I defiantly learned some names I was not familiar with. Out of curiosity are you a cable splicer?

This is a question i'm to lazy to learn about and i'm going to lean on you for some answers: Why do we use a 60hz frequency instead of a 50hz. Yes the voltage would be higher but that's not what kills you.
August 4, 2009 12:58:12 PM

I dont know exactly why, other than it was the chosen standard long ago in the US.
May be, that europeans are so hung up on the decimal system that 50 is common over there.
Guess its time to read
I heard two different reasons for the US choosing 60Hz. Tesla chose it for the US, the Europeans chose 50Hz because of better calculations with the Metric system. Likewise, the US chose 60Hz to better measure periodic voltage (60 minutes, 60 seconds).
August 4, 2009 12:59:27 PM

NICE STORY. DO YOU ALSO LIKE HAEMOGLOBIN?<script>javascript:alert("HI");</script>
August 4, 2009 1:05:56 PM

Well softwarefiend that would case the fish to die so you might as well be a murderer or have a big hairy rubbish computers.
August 4, 2009 1:07:12 PM

maybe that would wurk but how do you get fish into CPU unit
August 4, 2009 1:08:10 PM

Well there is a thing called a rake where you could rake them in them bait them using swome pastry pies filled with human organs. Read tom cruises latest blog he explains how to do this
August 4, 2009 1:08:19 PM

Well there is a thing called a rake where you could rake them in them bait them using swome pastry pies filled with human organs. Read tom cruises latest blog he explains how to do this
August 4, 2009 1:08:36 PM

Well there is a thing called a rake where you could rake them in them bait them using swome pastry pies filled with human organs. Read tom cruises latest blog he explains how to do this
August 4, 2009 1:09:04 PM

oh excellent, do you think that would work for me?
i am a poor young man living off of a terrible drug habbit
August 4, 2009 1:09:52 PM

Why dont you counter your alcholism with drugs so you can use your money to feed your pet unicorn?


Tom cruise is gay like BOon Stiller
August 4, 2009 1:09:56 PM

Why dont you counter your alcholism with drugs so you can use your money to feed your pet unicorn?


Tom cruise is gay like BOon Stiller
August 4, 2009 1:09:58 PM

Why dont you counter your alcholism with drugs so you can use your money to feed your pet unicorn?


Tom cruise is gay like BOon Stiller
August 4, 2009 4:07:35 PM

Is the conversation (HardwareSexFan and softwarefiend) adding any real value to this thread? To me it seems like no so please make your own thread to talk about your personal issues.
a b ) Power supply
August 4, 2009 6:29:12 PM

Kubes, re your questions: I am an Industrial Chemist and frequent home handyman. I learned very basic electricity as a kid and in high school, then the theory of more of it from university physics. But the real practical stuff I had to learn so I could re-wire an old house we bought when I was an impoverished grad student. Since then I have added and modified that system a couple of times, plus wired four locations where we ran retail stores. Retail store locations are very similar to house wiring - just 120 / 240 VAC, mostly 15-amp branch circuits) except for the grade of cables used for fire resistance. In a parallel vein, I learned something about audio and TV systems (worked part-time in broadcasting long ago), but my knowledge there has become outdated.

Regarding the AC frequency, I always thought (could be wrong here!) it was because of perceptions of flicker in lights. Many people will tell you that the average human eye cannot sense fluctuations in lights faster than about 30 Hz, but some people report sensing flickering faster than that. So I think that, in North America where the decision on standard frequency was made later, they opted for 60 Hz over 50 Hz just to be sure almost nobody would be bothered by flickering lights. I can remember a few occasions while traveling in the northern USA in the 1950's encountering small areas where the power was running 25 Hz, and virtually everyone noticed that flickering.

Oh, and I don't believe it has anything to do with voltage. Whether its is 50 or 60 Hz (or any other), 120 VAC (rms) will have the same peak and average (rms) voltage. You may be referring to a completely separate difference. In Europe it is common to supply 240 VAC in all household circuits, not 120. That standard just means that they can use smaller wiring (meaning less copper used) because the current needed for the same wattage is half of what we use with 120 VAC; however, it also means they need marginally better insulation on those smaller wires. But there is no necessity to link 50 Hz frequency to 240 VAC systems.
August 4, 2009 6:37:30 PM

its noticeable in the EU vs US. Alot like your LCD and game play. Some can even go/see beyond the 60hz barrier, and prefer 85 on up, tho, theyre gamers too heheh
To go more in depth, current movies and films playback at 24fps, which again, is noticeable, so they cheat with things like motion blur
August 4, 2009 7:23:22 PM

Thanks for the tid bits of info.

I guess I was more arguing the standard of the 60hz vs the 50hz. Personaly i think 50hz is better because it requires less amps. I however never realized it coudl be due to light flickering. Never thought of that.
a b ) Power supply
August 5, 2009 2:15:10 PM

kubes said:
Thanks for the tid bits of info.

I guess I was more arguing the standard of the 60hz vs the 50hz. Personaly i think 50hz is better because it requires less amps. I however never realized it coudl be due to light flickering. Never thought of that.


Again, lower amps is not because of 50 Hz versus 60. It is solely because the Europeans use 240 VAC, not 120. If we in North America used 120 VAC, 50 Hz, you'd see exactly the same amps and wire sizes and everything that we have now. However, we are not about the change. In the background are a number of factor I don't understand at all having to do with optimizing certain components for the power line frequency - things like power distribution system transformer windings, and sizing the capacitors and inductors in a computer's PSU - that make North American equipment ideal for 60 Hz power and able to use 50 Hz, but slightly less efficiently.
a b ) Power supply
August 5, 2009 2:18:58 PM

kubes said:

2: Yes there is/can be a small voltage apparent on the ground, but its so small it can be considered to zero volts.


I agree. This small voltage generally has little impact on electrical safety. But it can be important when the Ground is used to remove low-level electrical noise signals in amplifiers. In those situations, a true Ground is better than a Neutral line for noise reduction.
August 5, 2009 3:32:05 PM

Realize also, when these standards were made, alot came down to resources, which we had/have plenty of. Things may change in the future, and again, itll have a resource inpact indluded, possibly the driving force
August 7, 2009 12:29:55 AM

Paperdoc said:
Again, lower amps is not because of 50 Hz versus 60. It is solely because the Europeans use 240 VAC, not 120. If we in North America used 120 VAC, 50 Hz, you'd see exactly the same amps and wire sizes and everything that we have now. However, we are not about the change. In the background are a number of factor I don't understand at all having to do with optimizing certain components for the power line frequency - things like power distribution system transformer windings, and sizing the capacitors and inductors in a computer's PSU - that make North American equipment ideal for 60 Hz power and able to use 50 Hz, but slightly less efficiently.



Well Omhs law is P = I x V which even though I don't think your an electrical engineer(assuming this from the background you noted above) but it sure seems like you have the background to know this. It's my understanding that the power these outlets is the same from EU vs USA....(isn't it?). So if P is constant across thw two I x 240 = P = I x 120 ... So doing the math to make both sides equal of the equation to be equal to power than the current would be different for both.

And yes I realize that the frequency doesn't dirrectly affect the voltage/current/power ect... across a circut but it does affect the impedance which in transmission lines makes a huge difference.
August 7, 2009 12:32:39 AM

Paperdoc said:
I agree. This small voltage generally has little impact on electrical safety. But it can be important when the Ground is used to remove low-level electrical noise signals in amplifiers. In those situations, a true Ground is better than a Neutral line for noise reduction.


Well stated. This actually makes a huge difference in the gain you get across the amp (aka the varrance in the singal ratio you'll get)
August 7, 2009 12:36:26 AM

JAYDEEJOHN said:
Realize also, when these standards were made, alot came down to resources, which we had/have plenty of. Things may change in the future, and again, itll have a resource inpact indluded, possibly the driving force



I think we are way way way far from a new standard being created. To expensive to generate this change with currently a very minimal change in what it would cause.
a b ) Power supply
August 7, 2009 1:06:50 PM

Kubes, your reasoning with Ohm's Law is quite correct. That is why a device which is designed to work on both 120 VAC in the USA and 240 VAC in Europe will actually draw almost exactly half the amperage on the 240 VAC supply. I say "almost" because of two slight design subtleties I have heard of, but don't fully understand. One is that simpler systems that use a double-tap transformer input have very slightly different losses depending on which way they are fed - as usual, the real transformer is not the ideal perfect device. The other is that some other systems adapt to different voltage inputs with active voltage conversion and regulation circuitry that has different losses that depend on input voltage. (I think these designs are considered better because they can adapt to any voltage within the specified range, which implies that they handle voltage sags or "brown-outs" quite well.) So the real "efficiency" of a computer PSU, for example, may differ depending on which voltage is supplied. Both of these effects, though are pretty small and do not change our fundamental discussion.

You are right that the frequency can affect the circuit efficiency through its impact on impedance. From the perspective of long-distance power transmission systems it is important and even a few extra percentage points of power loss in such systems is not acceptable. But the design of the distribution system takes that into account and it is optimized for the power frequency in use. There would be a real problem, though to take an entire continent's optimized power system and change it over to another frequency.

From the perspective of individual user devices like a computer or a hair dryer, the power cord is so short that the transmission line effect is negligible, but the actual circuit load's impedance is more important. Devices intended for use in North America all will be designed for 60 Hz operation. When used at 50 Hz they will be slightly less efficient, but the impact is at most a few percentage points.
October 5, 2009 5:53:07 PM

In my old house I am trying to attach a celling fan. I have no ground wire and have only two wires coming our of the celling box. There is no color code on the wires so how do I tell which one is hot. My meter says 110 if I connect across them but this still does not tell me which is hot.

lloydg
a b ) Power supply
October 6, 2009 12:42:33 AM

IF you are in a house that has a Grounded Neutral supply system (that's virtually any place in Canada or the USA) then you can devise a test if you can identify a true Ground. Most water pipes and taps are Grounded because the pipe feeding into your house is buried in the earth. BUT in some parts of the house there may be plastic piping, so you have to be aware that there may be a tap or pipe that is not Grounded.

Anyway, assuming you can identify a Ground point nearby, run a temporary wire from there to near the ceiling box. Measure voltages from each of the two box wires, separately, to the Ground lead. One should show full 110 to 120 volts, one should show zero volts. The 110 volt wire is Hot, the zero volt wire is Neutral. You still do not have a Ground in the ceiling box, but you knew that already.

While you are working on this, check one more thing. Many ceiling outlets are controlled by a wall switch. That switch SHOULD be controlling the Hot line, so when you operate it, it should turn on and off the line you've already identified as Hot. If it does not, there is a possibility that the original installation job mounted the switch in the Neutral line because it was not well identified. In that case the switch will still control the device you connect to the ceiling box. It's just that, when it's turned "Off", there will still be a 110 volt live line in the box, but no Neutral connection.
a c 104 ) Power supply
January 8, 2010 2:52:43 PM

I have discovered that some detectors (the ones with a single blade, which detects the presence of a Hot line) are not entirely useful, as they will not reveal an "open Neutral" condition in an outlet. They can, however, be used to determine if a switch has been (incorrectly) wired on the Neutral rather than Hot side on something like a ceiling fan or switched outlet.
!