Can someone explain grounding to me?

Factual Questions
41

The typical house is wired from a split phase, meaning that there is a transformer outside that supplies 240 volts and the center tap of the transformer is the nuetral (which is grounded). Line to nuetral is 120 volts for either line, and line to line voltage is 240 volts.

It’s much less common, but there are still some places that supply two phases of a three phase system. In this case, instead of the 120 volt lines being 180 degrees out of phase with each other (which is what you get from a center tapped transformer) you end up with the 120 volt lines being 120 degrees out of phase with each other. The line to nuetral voltage is still 120 volts, but the line to line voltage is only 208 volts.

Typically you only find the 208 circuits in very old systems. The power companies will typically rip those old systems out and replace them as soon as they have a decent excuse to (like they need more current capability on one particular branch).

In a typical residential service, as one voltage peaks at 120 volts the other is 180 degrees out of phase, so it is peaking at -120 volts. You end up with 240 volts between them.

42

In case anyone is wondering where this comes from, in a three-phase system the L-L voltage is equal to SQRT(3) x V[sub]L-L[/sub] = 1.732 x 120 = 207.84 V ~= 208 V.

43

I know. A poster suggested they were in phase, and I was suggesting why they couldn’t be. :slight_smile:

44

Not to be tooting my own horn, but a certain SDSAB member has written a Staff Report on grounding on the space station.

45

Ok, I can see how it can be out of phase with respect to the center tap. The amplitudes however, are in phase timewise, as each reaches it’s peak value at the same time and is zero at the same time. It’s just not the same meaning of ‘phase’ that is used when refering to single phase v. three phase.

46

I spent a year researching electricity so that I could create a stage presentation explaining it to grade school children, so I’ll have a crack at this.

Lib, here’s as decent a starting point as any.

For centuries, electricity simply referred to what you could do with a variety of substances, such as, IIRC, a peice of wool and a glass rod. You rub the wool on the rod vigorously for awhile, and suddenly you can make the rod attract very small pieces of paper or straw as though it had suddenly become sort of magnetic. An object given this property was said to contain an “electric charge”. Many wondrous demonstrations were given (and still are) as to the amazing effects you can achieve with electricity.

People investigating this phenomenon discovered two important things:

  1. that there were two different and opposing kinds of charge, because you could use one type to cancel out the other type. At some point these were named “positive” and “negative” rather arbitrarily.

  2. that certain substances, known as conductors, can move a charge from place to place. What really happens is that charge will spread itself out over the entire surface of a conductor (such as a variety of metals), whereas a non-conductor keeps any charge it acquires at the acquires at the acquisition site. Our glass rod is only charged at the place where the wool was rubbed. Any two charged non-conductors connected by a conductor will have their charges equalized.

Later it was found that the carrier of negative electrical charge was the electron, and the carrier of positive charge was the proton. Electrons can move comparatively easily away from the atom they orbit (unlike protons), and it eventually came to be realized that an object with a negative charge has an excess of electrons, and that a positive charge simply meant the absence of electrons. It would have been much more intuitive if the electron charge had been randomly named “positive”, but we’re stuck with it the way it is, unless some world scientific body wants to engage in a lot of writing, retraining, and book burning.

The earth contains mostly atoms that are in electric equilibrium, i.e., they have as many electrons as protons (which have equal but opposite charges) and so is electrically neutral. Therefore, if you want to make sure an object that might gain an unwanted electric charge remains neutral, you hook it up to the earth by means of a conductor. Then, any charge it acquires gets equalized with the earth’s charge. It would take a hell of a lot of charge to make the Earth’s charge leave neutral, so what happens is the other charge is neutralized.

Our glass rod trick works because it is not a conductor (and you don’t touch the charged part of the rod while performing the trick). If you tried the trick with a conductor, the charge would even out over the surface of the conductor, meaning the charge would come into contact with your body, which is also a conductor. The charge would then spread out over your surface along with the conductor you hold, which, if connected to the earth (ground), would neutralize it. This, by the way, is why you never want to grab anything that has a charge on it unless you wear an insulator (non-conductor) like a thick rubber glove.

47

Pretend you have one wire that has 120V peak, and the voltage at any time is given by
sin x.

If you had another line with 1V peak on it in the same phase, it would be represented by the same formula: sin x.

What is the difference between the two?
sin x - sin x = 0.
They have to be 180 degrees out of phase, so that you end up with
sin x - sin (x+180). But sin (x + 180) is just -sin x. So the difference between the two is
sin x + sin x = 2sin x.

Since our peak voltage is 120 volts, that means the potential between the two has a peak of 2 X 120 or 240 volts.

One has to go down while the other goes up or you’ll not be able to get more voltage.

48

The deal with moisture in the ground, and water and electricity not mixing well in the first place, is that water is a polarized molecule. The oxygen atom, being electrically stable but wanting to fill out its outer electron ring, shares two electrons with two hydrogen atoms, which have one each. The electrons stay mainly nearer to the oxygen nucleus, which means that 10 electron (negative)charges overbalance the 8 proton (positive) charges, and in the hydrogen nuclei, the “missing” negatives are overbalanced by the proton in each.

Because the water molecule arranges itself in such a way that the hydrogen atoms form almost a 60 degree angle to each other through the oxygen atom, water basically ends up with a positive end and a negative end, meaning, for example, that the positive end can attract to a negative end of another water molecule, which can cause a body of water to act as a conductor.

If you connect a charged conductor to water, you effectively extend the conductor, but in a basically unpredictable way, given the non-linear shape of the water molecule and the ways they can connect electrically. A little bit of moisture in some soil, as long as it is mostly composed of non-polarized molecules, will not pose much of a problem, but you wouldn’t want to ground anything to really damp mud or an actual puddle, or you may not actually be grounding at all.

49

As I hope my posts have explained, nothing gets a shock until two entities with different amounts of charge are connected through a conductor.

Three things then come into play. The voltage, the resistance and the current.

First you need a unit charge, which is called the coulomb, having been dreamed up before the discovery of the electron. The amount of charge on an object is measured in coulombs.

Voltage is a measure of elctric potential energy, measured in Volts, which are Joules of energy per coulomb of charge. This is potential energy, representing basically a capacity for action.

Resistance is the ability of an object to impede the motion of charge (called current) across it. Some substances like rubber, resist current very well, whereas others, like some metals, resist it very poorly. This is the distinction between conductors and insulators. Charge passing through a resistor uses up energy, meaning the voltage on either side of a resistor is different, assuming the tow sides of the substance are connected to different charges.

A vacuum has an infinite resistance, while air has a very high resistance but under extreme conditions can act as a conductor.

Current (the speed of the spread of electric charge, measured in coulombs per second past any given point) is what happens when a voltage meets a resistor. The higher the the ratio of the voltage to the resistance, the faster the current flow.

“Shock” is caused by current flow.

“Closing a circuit” means to connect two differently charged objects with a conductor.

If you walk across certain carpeting on a very dry day, you will build up a charge on your body, which will not be dissipated becuase of the insulating qualities of your shoes relative to the carpet or by the connection of stray water molecules in the air. When you touch a friend or a doorknob, which are most likely neutral, you are connecting two different charges through a conductor, current flows (the greater the difference in charge, the faster the current), and if it’s fast enough, you may feel a shock. If the differences in charge are very great, you may even see the air between you and what you touch start conducting, meaning you see a blue-white flash.

50

Right. I understand all that. I was just pointing out that 240v home service was still just single phase. That is, you could run a 230v single phase motor just fine on that service, but not a 230v three phase motor. It’s getting off topic at any rate, so I’ll shut up now.

51

Yes.

52

There are a lot of valid reasons for ‘grounding’ and most have been described here…but unless I missed it one of the major ones hasn’t been addressed yet; I’ll call it “neutral float”.

In a ‘typical’ US home Let’s say you have the black (‘Hot’) and white (‘Neutral’) wires delivering 120V to an appliance. Then there is also the green (‘Ground’) wire connected to the case or something of the appliance. Back at the service entrance panel the green and white are both connectred and both go to the rod going into the Earth. If you put an AC voltmeter across the white and green wires, at the service entrance, you’d get a reading of zero.

However, out at the user’s end of the wire, by the appliance, things are different. Since the appliance draws current, there is a voltage drop in the white wire due to its unavoidable resitance. The more current, the bigger the voltage. If you connect an AC voltmeter across the green and wires at the appliance, you could read several volts for a ‘hefty’ appliance.

Now, if something goes wrong and the white wire touches the case or green wire, that voltage will cause a lot of the load current to flow back through the green wire back to the service panel, hopefully with no problem.

However, if you didn’t have that green wire, and the white wire touched earth through a person or object, bad thinks could happen; a common one would be sparks fromn the high currents involved, which can cause fires (think inside a wooden wall) or even explosions in some environments (think of a gas station).

Many folks have discussed the disaster that could happen if the ‘hot’ wire touched someone or something that also touched ‘earth’. The above discussion attempts to explain that there are also significant problems even if the Neutral were to touch ‘earth’, and that the green ‘ground’ helps reduce that risk.

At least - that’s as I recall it from EE class.

53

bobk2 You seem to be recalling EE class incorrectly.

Typical house wiring has the white wire connected to ground at the breaker box.

At the plug the voltage from white to ground should be zero. The voltage from white to black should be the same as the voltage from black to ground.

There won’t be “several volts” between the white and the ground at a hefty appliance, even if the appliance is drawing a strong current. why? The resistance of the white wire from the breaker box to the appliance is the same as the resistance of the ground wire.

The main purpose of the ground wire is to bring a redundant ground to the appliance. The white and black wires power the appliance. The ground wire shields it.

Consider an electric bench motor. From the plug, the white wire and the black wire connect to the motors terminals. From the plug, the ground wire is connected to the metal motor housing. What that means is that the exposed parts of the appliance (the motor’s housing) have their very own dedicated connection to ground. If your skin touches the motor housing while you lean on a water pipe (ground) nothing happens. **(A) **If the shielding inside the motor breaks and the white wire comes in contact with the motor housing it may not be noticable since the motor houseing has the same potential as the white. **(B) **If however, the black wire loses its sheild and touches the motor housing it will immediatly fault the circuit and trip the house breaker.

Now, replay the whole scenario without a dedicated ground on the motor housing. **(A) **White wire touches housing and, like before it now grounds the housing, so when you lean on the water pipe and touch the housing nothing happens. **(B) **But if the black wire touches the housing of a motor that wasn’t grounded it makes the housing equal to the hot wire. Then you with your waterpipe will get zapped.

GFCI circuits take that one step further by monitoring the ground wire for any signs of current and opening up (GFCI devices will react to case A listed above, while standard circuit breakers won’t. That forces your appliances to strictly adhere to separation of operating ground and mechanism housing, an added precaution)

54

[QUOTE=BubbaDogGFCI circuits take that one step further by monitoring the ground wire for any signs of current and opening up…[/QUOTE]

Ooooo…you almost had it. GFCIs do not monitor current on the ground wire. They will, in fact, work even on ungrounded electrical systems. A GFCI works by looking for a curent imbalance between the current in from the hot wire and the current out through the neutral. It does ths by passing both the hot and neutral through the center of a current transformer. If the current in equals the current out, then no voltage is induced across the CT secondary (we say the currents “buck out”). However, if some of the current is diverted to ground, say, through you, a voltage proportional to the current difference is induced across the secondary, and if the current imbalance is 5 mA or more, the GFCI trips off.

55

In my post I said the white and green are connected together and to earth ground at the service entrance panel which the statement above does not contradict. I’m using “service entrance panel” to mean “breaker box” and perhaps some folks will take it to mean something else, but for the purposes of this discussion it is accurate.

Also, I just went out and looked at my breaker panel. It is as I said. All the green (bare, actually) and white wires from all over the house are connected together along with a very large white wire coming from the meter (Neutral on our 220V service) and also to a very big green wire which looks like it runs out to earth ground (I can’t quite follow it all the way)

I haven’t broken out the multimeter yet to verify that at the appliance the green goes to the case and the white doesn’t, but I suppose I could if needed.

“Volts” of float might be a little extreme in a house, but not impossible, but it is very possible in an industrial setting.

So, for the moment at least, my EE recollections seem (to me anyways) to be correct.

56

This isn’t true, because under normal conditions the ground wire carries no current. If there is no current, then there is no voltage drop. If there is current (as in the white wire) then there is a voltage drop.

Consider a “hefty” appliance that draws 20 amps of current, and for the sake of some easy math let’s assume the wire has 1/2 ohm of resistance (which is not an unreasonable number). The voltage drop across the wire is going to be the current multiplied by the resistance, or 10 volts. You’ll actually have a 10 volt drop across the black wire as well, since it’s made out of the same length of copper, and only 100 volts actually across the appliance itself (120 - 10 - 10).

Granted, 10 volts is a lot less dangerous than 120 volts, but the point is that it isn’t zero.

You have a couple of choices now. You could not bother to connect either wire to the case. This would be bad, for the reasons mentioned, since a short from the black wire to the case would leave the case at 120 volts. You could connect the white wire to the case, but this potentially puts the case at 10 volts above ground, so if you are touching the appliance (maybe it’s a stove/oven) and something grounded (like water in the sink) you could get a shock, so that’s not a good answer either. Hence, we run a seperate wire that carries no current (the green wire) and attach the case to it. You still have a potential problem of what happens if the white wire touches the case, but then you have current flowing through the ground wire, and that’s what GFCI’s are for.

57

You may already know this, but don’t make this measurement with the appliance plugged in. Multimeters measure resistance by applying a small voltage and measuring the resulting current, which it expects to be relatively low. If there is a current present, it will definately screw up the measurement and may possibly damage the meter.

The green wire is supposed to be connected to the case. If the white wire is connected to the case then you have a problem.

58

Yes, if this really got down to brass tacks I was going to unplug the electric dryer and make my measurements there. But again, I’m quite confident its as you (and I) said, without having to measure.

59

Ya got me.

I should never post at work while trying to answer six questions on four subjects.

I misread bobk’s statement and took off on another tangent. And ** Q.E.D.** called me out for oversimplifying the operation of a GFCI.

I’ll just whack myself with a sliderule and go sit with the accountants.

60

Let’s get voltage magnitudes between green ground and white ground straight. ** QED** and bobk2 have it right. There should be zero, nihil, nada current in the green wire and so it will be at the same voltage throughout its entire length as is the white wire at the breaker panel where the two are connected together. The white wire out at the load end will be at some voltage above that depending upon the current and the length of wire between load and panel.

According to my wire tables #12 wire has a resistance of 1.588*10[sup]-3[/sup] Ohms/ft. This size wire is allowed 20 Amp. by the code. If the wire is 100 ft between load and panel and carrying maximum allowable current, the white “ground” at the load will be about 3.2 v. above the green ground voltage.