Understanding Electrical Ground & How It Works

[Mårten]

Thanks for patience and all answers...

Charges in a conductor will move as far away as possible from each other, but this is not really relevant to the flow of current in a wire. Current flows in a wire because there exists a difference in potential between one end and the other. A basketball rolls downhill for the same reason.

But wait for a moment here... What is potential? Isn't that electric potential energy per unit charge? And a high magnitude of the potential energy for a specific charge (e.g. +) (relative to some point at some distance) means that you have a charge very close to a gathering of some other + charges, and that the electric field between your charge and these other charges then is strong (pointing away from each other). And therefore your charge will run away fast. So this will create a current (as long as we also have some EMF continously filling up new charges in the charge gathering). At least, this is my theory now of how it works after been studying the answers in this thread thoroughly. Or have I missed something here with the concept of potential?

 

[atyy]

What is potential?

Let's consider again my favourite example with capacitors.

Let's put 1 + charge on plate A, and 1 - charge on plate B. They would like to be with each other. In circuit theory, they can't get together because the two plates are not connected by a metal wire (outside of circuit theory, if you bring the plates close together enough, or put zillions of pairs of +/- charges instead of just 1 pair, then the charges will express how much they like to be with each other by jumping across the plates without a metal wire connecting them, and you will see this as a spark).

Anyway, in circuit theory, the charges would like to be with each other. In more technical language, we say that the separation of charges creates an electric field between the plates, and the electric field indicates the direction in which the charges would move if the plates are connected by a wire. The important thing to note is that the electric field is present in between the plates where there is no charge present.

An electric field is a *vector* field, which we imagine as an arrow attached to every point in the space between the plates indicating the direction and magnitude of the electric field. In general, the arrows at different points can have different directions and magnitudes. By some magic, it turns out we can calculate all the right answers if we replace the vector field with a *scalar field*, which we imagine as only a magnitude - no direction! - attached to every point in the space between the plates. This scalar field is called the electric potential. The electric potential, although it is just a number, encodes the directional information in the electric field in the following way: the direction in which charges move is in the direction of lower potential, ie. down the potential difference. So now you can kinda see why only potential differences have meaning, whereas potentials themselves don't - if you raise the potential at every point between the plates by the same amount, the potential difference between any two points remains the same, and hence the directional information remains the same. Anyway, the point here is that just as there is an electric field between the plates where there are no charges, there is an electric potential between the plates where there are no charges.

In almost all physical situations, every electric field or electric potential difference is ultimately created by some charge separation. However, since the field or potential can exist in regions where there are no charges, it makes sense to consider them as the actors in those regions. This is why chroot said that the electrons are moving down a potential difference.

One minor point, the magic of replacing the vector electric field with the scalar electric potential only works when you have steady currents. Strictly speaking, this magic doesn't work when you are charging a capacitor, or have alternating current - however, the approximation of an electric field by an electric potential remains very good as long as the current doesn't change too quickly.

 

[schroder]

The Earth is a very poor conductor. It is used as a reference for voltages, NOT as a conductor.



Integral mislead you in his previous post. You can get current to flow from the hot conductor to the ground conductor in your household wiring because the ground conductor is tied to the neutral conductor in your breaker box. The power company does not pass current through the earth.

- Warren

These statements are not quite correct. The power generated by the electric company is always referenced to earth. The Earth is a very good conductor of electricity! The resistance of any conductor varies inversely with the cross sectional area of the conductor. What has a bigger cross section than the earth? In the early days of power generation and distribution, no ground or neutral wire was used in the transmission system. Sometime later it was determined that not all Earth references at the distribution centers had equal ground potential and certainly not all final users (homes) had equal ground potential which caused an unequal and very dangerous distribution of voltages. So all national electrical codes mandated that the power companies shall provide a separate neutral lead to ensure the equal distribution of voltage to all end users. But at all large commercial sites and power stations the National Electrical Code still mandates that a separate “safety ground to earth” be established and maintained at a resistance of not greater than 25 ohms. Sometimes electrodes need to be driven deep into the Earth to achieve this , but it can be achieved just about anywhere on the earth.

 

[atyy]

Isn't that electric potential energy per unit charge?

Yes. So back to the capacitor with 1 + charge on plate A and 1 - charge on plate B. Let's assume that they are confined to those plates (no sparks). There is an electric field between the plates, ie. if you put another + charge near plate A (let's call this the test charge), it will be repelled from the + charge on on plate A, and will be attracted to the - charge on plate B. If the test charge is allowed to move in the electric field from plate A towards plate B, the electric field will be doing work. Since we can describe an electric field by an electric potential, we can also say that the electric potential is doing work.

Since potential is arbitrary. Let's set the potential at plate B to be zero.
Let the potential at plate A be V.
So the potential difference between the plates is V.
Let the test charge be q.
Then the work done by the electric field/potential in moving the test charge from plate A to plate B will be qV.
So the work done per unit charge will be qV divided by q = V.
That's why the electric potential is the electric potential energy per unit charge, as you said.

Minor point: There's a conventional double minus sign, which I'm ignoring.

More important: I've assumed that the + charge on plate A really is stuck there, that it doesn't move towards - charge on plate B although it is attracted to it. I've also assumed that the placement of a test charge near the + charge on plate A doesn't cause the + charge on plate A to move, even though it is repelled by the test charge.

 

[chroot]

The Earth is a very good conductor of electricity!

The Earth is a very poor conductor of electricity when compared to, for example, a copper wire. Your statements later in your post regarding differences in ground voltage are a good example of how poor a conductor it is.

- Warren

 

[schroder]

The Earth is a very poor conductor of electricity when compared to, for example, a copper wire. Your statements later in your post regarding differences in ground voltage are a good example of how poor a conductor it is.

- Warren

You need to clarify your statement. If you are talking about a cubic meter of “earth” as compared to a cubic meter of copper then certainly the copper is a much better conductor. However, when the term “earth” is used in reference to the Earth itself, it is the best possible conductor of electricity because of the cross sectional area, as I pointed out. It is necessary to establish a proper connection to this earth, however. Sometimes that requires driving electrodes deeply into the ground, or possibly treating the ground with chemicals or water. But once a good gound is established it is very possible to transmit electricity from one place to another without the need for a separate ground or neutral wire. As I have already mentioned, that was the practice in the early days of power transmission but because the ground references at the various distribution points were not all properly done and verified it soon became apparent that a separate neutral wire must be provided. I suggest you do some research on this.

 

[stewartcs]

You need to clarify your statement. If you are talking about a cubic meter of “earth” as compared to a cubic meter of copper then certainly the copper is a much better conductor. However, when the term “earth” is used in reference to the Earth itself, it is the best possible conductor of electricity because of the cross sectional area, as I pointed out.

The resistance of the ground rod to the Earth is also a function of the soil's resistivity as well as the length and radius of the ground rod.

But once a good gound is established it is very possible to transmit electricity from one place to another without the need for a separate ground or neutral wire.

Although this is quite true, it is not how electrical distribution systems are really intended to work under normal conditions. The Earth ground is mainly used when fault conditions occur. Utility companies don't depend on fault current to transmit power.

There is also a slight benefit to the grounds although it is not really an intended design per se. The repeated grounding points throughout the electrical distribution system (power transmission lines for example) help reduce power line losses by essentially creating a parallel network with the transmission line.

CS

 

[schroder]

Although this is quite true, it is not how electrical distribution systems are really intended to work under normal conditions. The Earth ground is mainly used when fault conditions occur. Utility companies don't depend on fault current to transmit power.

There is also a slight benefit to the grounds although it is not really an intended design per se. The repeated grounding points throughout the electrical distribution system (power transmission lines for example) help reduce power line losses by essentially creating a parallel network with the transmission line.

CS

Yes, I agree with this. I did not mean to imply that today’s power transmission systems use Earth ground to transmit power. I mentioned it in an historical context as well as a physics context to establish that it is possible. Besides the indirect benefits you mentioned, the main advantage today of proper grounding is the safety function, which I would like to point out by a specific example.
Suppose you are supplying three-phase 4000 Volt ac supply to a machine and the source and feeder resistance is 15 ohms. Now one of the phase wires in this machine touches to the motor frame which has a 10 ohm resistance to Earth ground. This will place about 2,300 volts on the frame of the machine with reference to ground, resulting in a flow of 92 amperes between the machine frame and ground. Now if a technician should happen to come along with his test cart, which is itself grounded, and touch the frame of this machine he will be subjected to roughly 920 Volts, which is more than sufficient to kill him instantly.
But if the machine was better grounded, say with a resistance to Earth of only one ohm, the voltage shock would be reduced to about 92 Volts and he will most probably live to correct the fault. There are also in place ground fault circuit breakers, but I am not taking those into consideration in this discussion. Proper grounding is one of the least appreciated and most misunderstood disciplines in electrical engineering and a thorough discussion of this would involve many more posts by myself and others.

 

[stewartcs]

Yes, I agree with this. I did not mean to imply that today’s power transmission systems use Earth ground to transmit power. I mentioned it in an historical context as well as a physics context to establish that it is possible. Besides the indirect benefits you mentioned, the main advantage today of proper grounding is the safety function, which I would like to point out by a specific example.
Suppose you are supplying three-phase 4000 Volt ac supply to a machine and the source and feeder resistance is 15 ohms. Now one of the phase wires in this machine touches to the motor frame which has a 10 ohm resistance to Earth ground. This will place about 2,300 volts on the frame of the machine with reference to ground, resulting in a flow of 92 amperes between the machine frame and ground. Now if a technician should happen to come along with his test cart, which is itself grounded, and touch the frame of this machine he will be subjected to roughly 920 Volts, which is more than sufficient to kill him instantly.
But if the machine was better grounded, say with a resistance to Earth of only one ohm, the voltage shock would be reduced to about 92 Volts and he will most probably live to correct the fault. There are also in place ground fault circuit breakers, but I am not taking those into consideration in this discussion. Proper grounding is one of the least appreciated and most misunderstood disciplines in electrical engineering and a thorough discussion of this would involve many more posts by myself and others.

I agree that grounding (not to mention bonding) is not very well discussed in Electrical Engineering courses - most unfortunate.

CS

 

[transgalactic]

we were told in our starting course of EE
that electrisity as we see it at home
has nothing to do with EE

further more we were told that
todays electrical engeneer doesn't have to know what's triple faze current

 

[stewartcs]

we were told in our starting course of EE
that electrisity as we see it at home
has nothing to do with EE

further more we were told that
todays electrical engeneer doesn't have to know what's triple faze current

That's most unfortunate that you were told that because it is not true. Two of the EE's that work with me specialize in power systems. If they didn't know anything about three-phase systems they wouldn't be employed.

CS

 

[Mårten]

For anyone who passes by this thread, I'm adding the following helpful link which sorts things up and explains quite a lot about grounding:

http://www.ese.upenn.edu/rca/instruments/misctutorials/Ground/grd.html

(Thanks Redbelly for the tip!)