Isolation transformer and electrons

[ArthurDent]

IanB – I don’t see where anything I have posted has been incorrect in my posts describing the proper use of grounding, getting shocked, or correct wiring principles. Please look at the last diagram I posted. The diagram of the Isolation transformer I show and showed previously in post #20 is exactly what Hero999 later showed to demonstrate whether you get shocked or not on primary or secondary of an isolation transformer-there is no disagreement there at all. I clearly show that touching ground and a hot wire referenced to ground is dangerous but that has nothing to do with the correct circuit return being a wire as universally used.

It is the continual claiming that ground is used as the return current path for an electrical circuit, not just for safety, that I see as absolutely wrong. Hero999 even linked to rarely used single wire feed with earth return so using earth for the return circuit instead of a wire is what they meant and are presenting, not a grounded person accidentally touching a live wire and getting shocked.  I have presented links to technical references to show that using a return wire and not relying on ground is universally accepted and I don’t see any disagreement with the references.  I also don’t see what I have posted as condescending, instead  saying that I’m silly and wrong without being able to explain how anything I have said is incorrect is what I see as somewhat condescending.

So simply put it is this. I have continually said and shown that a person can get shocked by touching a live wire that is referenced to ground. Probably all of us have gotten shocked at one time or another. I have also said that in an proper electrical circuit there is a neutral wire that is also bonded to ground (why else would you use an isolation transformer?) and it is the wire, not the somewhat resistive ground, that is the circuit return path back to the source. How is that in any way incorrect?

[IanB]

Nothing you have said is incorrect, but neither have a seen a "continual claiming that ground is used as the return path for an electrical circuit", so there seems no need to correct that. There has been perhaps a misleading drawing, and a reference to the rare use of single wire earth return, but this is not the same as "continual claiming". On the whole everyone just seems to be violently agreeing with each other.

Anyway, I'll let it rest.

[cdev]

Some power systems in Australia, New Zealand and elsewhere use ground for returns. Because of the lower cost. But not many.

https://en.wikipedia.org/wiki/Single-wire_earth_return

There is a history of power systems in Australia on the IEEE site that goes into a lot of detail about the systems' evolution there.

[Zero999]

Post #40 Hero999 – “In the distribution transformer scenario, the earth simply acts as a return conductor for the current path.”
Post #46 Hero999 – “I did read your posts, but they refer to distribution transformer, rather than the original poster's set-up.
As in your post #40 I am indeed talking about the distribution transformer scenario, as you were.

Hero999 – “The original poster is also in Europe and you refer to a US distribution system which is very different.”
Whether it’s the U.S. or Antarctica they still use wires to make a circuit. They do not rely on ground for a return path, only for safety. Back in the 1800s for telegraph (low current, low voltage) they had used it but switched. Rarely, in extremely high voltage long distance transmission, ground has been used as a return where the resistance of the ground and the current used is relatively small compared to the voltage. It has been tried and kind of works in these special applications. What was being discussed here was a medium voltage neighborhood distribution system where there is a physical wire used as the return and the ground is just part of the protective or safety circuit, not a return path as you indicated in your drawing. If the ground had any typical resistance at all you would never get full voltage to the load.

Sigh. . .
I've never said that the earth is routinely used as a path for the return current on low voltage domestic installations. You've has completely missed the point and taken the thread off topic by being the first to mention the distribution transformer. I apologise for making matters worse by replying to you and mentioning single wire earth return, which as you've correctly stated, is only applicable to high and medium voltage distribution.

Also in your drawing you show a 3-phase system which is mainly used for larger industrial applications and not residential.

You're wrong: a three phase system is routinely used both for domestic and industrial distribution in Europe. Single phase distribution transformers are very rare. We don't have many pole mounted transformers either. One three phase transformer will typically serve a whole street or block. In a domestic, single phase setting, every third house in a street might be on the same phase. For example in a street with 9 houses, numbers 1, 4 & 7 might be on phase 1, houses 2, 5 and 8 on phase 2 and houses 3, 6 and 9 on phase 3.

As far as an isolation transformer working I see that you agree with my previous diagram from post #20 on that point.

Yes, it illustrates why a meter will read 230V between live and earth, when connected directly to the mains but will read 0V between the secondary and earth on an isolation transformer. It demonstrates that when the transformer's secondary is earthed, there's a possible return current path through the earth, yet there is none when the secondary is floating. The ground current path is shown between the meter and transformer's earth connection. I repeat it's not supposed to illustrate the correct way to measure the mains, just that the full mains voltage will be present if one of the meter's probes is earthed and the other connected to live. Note that the isolation transformer's primary is connected between live and neutral, not earth!

[nForce]

So electrons also alternate in the earth, when conducting in the loop?

So again defining the parameters, if there is AC voltage potential between two points then you're ready for current to flow. By completing the circuit or loop it will conduct and at the electron level they will alternate directions at the frequency of the mains.

Like I said previously (unless you want to dive deep) just consider the earth to be a wire and if it's in circuit (and not isolated) electricity will conduct.

If we take an ampermeter and connect it in the earth let say 10 meters below ground, then we will measure AC current if we have single wire earth return power system?

[IanB]

If we take an ampermeter and connect it in the earth let say 10 meters below ground, then we will measure AC current if we have single wire earth return power system?

If you have an ammeter, it has two wires. You are going to stick one wire into the ground? Where are you going to attach the other wire?

Where does the single wire earth return transmission system come into the picture? Is it part of the test or is it far away?

[nForce]

We should be in a cave, below ground. One wire goes on one side of the wall the other goes on the other wall. Because of power grid and power stations nearby or above ground, we should measure AC current.

[IanB]

Yes, those are called stray currents. They always exist. Occasionally they are harmful and can cause galvanic corrosion on buried metal objects. But mostly they will be negligible.

Stray ground currents are why the London Underground has two insulated power rails and does not use the running rails to return the traction current.

[Zero999]

We should be in a cave, below ground. One wire goes on one side of the wall the other goes on the other wall. Because of power grid and power stations nearby or above ground, we should measure AC current.

Like this?

Yes, some of the current will flow round the cave and some through the meter.

As mentioned above, they can cause problems, if the system isn't implemented correctly, especially if the ground isn't conductive enough, resulting in a high potential difference gradient, near the transformer.

[nForce]

Why didn't you tell me about this at the beginning?

It took 3 pages of this topic just to explain to some beginner that yes an AC current flows in the earth because of the power grid and it's called a stray current.

I didn't know about this, because in convential circuits there is always a return wire. The earth does not look like a wire.

[Zero999]

Why didn't you tell me about this at the beginning?

It took 3 pages of this topic just to explain to some beginner that yes an AC current flows in the earth because of the power grid and it's called a stray current.

I didn't know about this, because in convential circuits there is always a return wire. The earth does not look like a wire.

Why did it take three pages for you to make it clear what you were asking?

Go back to your original post.

I have a basic question:

When we use isolation transformer we have a physical electric isolation between two systems. If we connect a wire between a system on the other side of an isolation transformer and ground then the current would not flow.

But how do electrons know on the other side of an isolation transformer that we don't have a current loop? They just feel a potential difference, so they should flow because they don't know nothing about isolation.

Thank you.

How was anyone supposed to know you were asking about stray currents? All you talk about is an isolation transformer and electrons?

Consider being more descriptive and posting some schematics, rather than a few cryptic statements.

[IanB]

Why didn't you tell me about this at the beginning?

It took 3 pages of this topic just to explain to some beginner that yes an AC current flows in the earth because of the power grid and it's called a stray current.

I didn't know about this, because in convential circuits there is always a return wire. The earth does not look like a wire.

Except...

Now imagine the Earth as a big ball of wire floating in space.

The current flows through you, through the Earth (a big ball of wire), through the rod in the ground, and back to the transformer.

[Brumby]

Why didn't you tell me about this at the beginning?

As already stated - but I will rephrase:  To get good answers you need to ask good questions.  We aren't mind readers and if all we know of you is what is in one or two posts, then we need you to help us understand exactly what it is you are asking.

It took 3 pages of this topic just to explain to some beginner that yes an AC current flows in the earth because of the power grid and it's called a stray current.

That is a rather simplified statement and not properly qualified.

Stray currents are unintentional, but occur as a function of the physics of a situation.  Fault currents are caused by a safety earth at the customer's location having a path to Live.  In a single wire earth return system the current through the ground in normal operation is, by design, the full current of the load.

I didn't know about this, because in convential circuits there is always a return wire. The earth does not look like a wire.

You are being too literal.  The Earth may not look like a wire, but it certainly DOES look like a conductor when you look at it properly - and this is the trick in understanding a few things with electricity ... what is a conductor may not seem like it should be.  Breathe on a high impedance circuit and the moisture that condenses from your breath will affect that circuit.  The human body does not look like a wire, but it conducts electricity - just hold the probes of your multimeter while on a resistance range.  This is also why electrocution is a risk.

[nForce]

Ok then, thanks for your help.

I have one question: Our electrical outlets at home are voltage sources. Because the voltage is constant and the current varies. But what are limitations that outlets would be current sources?

[Zero999]

Ok then, thanks for your help.

I have one question: Our electrical outlets at home are voltage sources. Because the voltage is constant and the current varies. But what are limitations that outlets would be current sources?

With a constant current source, the loads would need to be connected in series, for the current to be constant. If the loads were connected in parallel, the current would vary, depending on the loads, similar to connecting devices designed for operation off a constant voltage source in series. The open circuit voltage of a constant current source is theoretically infinite, with the power being proportional to the resistance, so open circuits need to be avoided and rather than disconnecting a load, it should be replaced with a short circuit, which will result an relatively little power dissipation.

[rstofer]

This question of transformer return current is easy:  The electrons have to complete a circuit back to the transformer from which they came.

Utility transformers have one side grounded and called the Neutral.  Every ground fault has to get back to this neutral point to complete the circuit and it does it by traveling through some path (perhaps dirt) back through the point where the neutral was grounded.

This is actually easier to think about when you consider a 3 phase transformer in a Y connection with the center point grounded.  A phase to ground fault MUST get back to that center point by any means necessary.  The resistance of the round trip path partially determines how much current will flow.

On some higher voltage installations, that midpoint is not grounded with a piece of wire but rather with a resistor.  This is done to limit the available fault current and is quite common in motor control centers for large motors.

On other, older, installations, the mid-point isn't grounded at all.  This means the first ground fault won't trip the circuit, it just guarantees that the other phases are at some high voltage relative to ground.  We used to have neon lights on the phases that, when one went out, indicated the presence of a ground fault somewhere in the system.  This type of installation is no longer in use.  The reason is that the second ground fault (assuming the first wasn't cleared) is now a phase-to-phase fault and the fault current is much higher.

[rstofer]

I think you might find some 1-wire medium voltage installations in rural areas of the US.  Apparently, New Zealand and Australia are also using 1-wire distribution schemes for rural electrification:

http://www.nzdl.org/gsdlmod?e=d-00000-00---off-0hdl--00-0----0-10-0---0---0direct-10---4-------0-1l--11-en-50---20-about---00-0-1-00-0--4----0-0-11-10-0utfZz-8-00&a=d&cl=CL1.18&d=HASH01a419b0834cf9dbfb8643d0.9.fc

[rstofer]

Four simple equations would suffice to describe all of these situations:

https://en.wikipedia.org/wiki/Maxwell%27s_equations

[Brumby]

I think a lot of that might be a bit too advanced for the OP just yet.  We are just getting sorted out on the fundamentals of current paths.

I have one question: Our electrical outlets at home are voltage sources. Because the voltage is constant and the current varies. But what are limitations that outlets would be current sources?

If we are talking about household general purpose mains outlets, the big issue here is practicality.

For a constant voltage supply, you present a given voltage at the outlet and the devices that get plugged in will only draw the current they need.  Aside from efficiency factors, there is no inherent waste of power.

For a constant current supply, the source will attempt to push the same amount of current through a device, irrespective of its needs.  As a result, the voltage presented will rise or fall as the case may be in order to maintain that constant current.  For a known load having a specific current requirement that does not change during operation, this could work - but for a constant current supply that wants to supply more current than needed, there will need to be a dummy load to "bleed off" the extra current that the load does not want.  Very wasteful.

If there is any fluctuation in the current requirements, then the source is going to "throttle back" when current demand rises and "open up the throttle" when it falls in order to keep the current constant.  This becomes a practical problem in many cases - such as starting and running an electric motor.  Have the supply provide the current for running and it will not provide the necessary current to get it going from a standing start.  Have the supply provide the current for starting and it will want to continue to supply that same current during the running period.  The voltage response will skyrocket.

Another problem would be with a power board.  Have a general purpose outlet provide a set amount of current and you plug in one device.  Let's assume that this setup works because the current supplied matches the current required.  Now plug in a second device with the same requirements (for simplicity).  That constant amount of current will be shared between two devices, so each will only get half.  Most devices will not work properly, if at all, in such a situation.  Fill a 4 way power board and it gets even worse.

But more than that the question has to be asked: Where do we place the constant current regulator?  Do we place one at each power point in a house?  Do we place it on each power circuit in a house?  Do we place it on the mains supply to a house?  Do we place it on the pole transformer that services several houses?  With the variations in loads, there is no real answer you can give here.  It's the power board problem all over again - at a bigger scale.

The only way around this is to have the constant current supply programmed to adapt to each load independently.  This would require the electronics to provide a constant current at every distribution point - and that electronics would need to be cheap, reliable and able to handle variations across a wide range of devices from a night light to a vacuum cleaner.  There would also need to be a communications protocol between the device and the supply so that the supply knows what current to deliver.

This is all getting way too messy.

What is needed is a system where the devices get to choose what current they require, without having to go through some convoluted process.  A system where adding extra devices through a power board, for example, just works.  A system that is simple, reliable and inexpensive.

Such a system does exist: Constant voltage supplies.  Present a given voltage and the device will only take the current it needs.  Paralleling loads is not a problem - so long as the capacity of the circuit is not exceeded.  There is no need for electronics in basic systems.  Reliability is high and costs are low.


The only time constant current supplies are generally beneficial is where you have a single functional unit - such as an LED - that operates more effectively and safely with such a supply.  This is one of the key functions of an LED driver.

[IanB]

There is one case where constant current circuits have been used, and that is lighting circuits. In particular, street lighting (in the past), and airfield lighting (maybe still today).

In the case of street lighting there would be one long loop containing many lamps in series, and a constant current power supply designed for a specified current with a large compliance voltage (maybe 1000's of volts). Each lamp used in such a circuit is specified with two numbers: the rated current, and the output lumens.

Because the lamps are in series, there are certain challenges with this system. For instance, if a bulb blows it will break the circuit and all lights will go out. Unfortunately, if a bulb blows the full compliance voltage will also appear across the break in the circuit, leading to a bright, hot arc with lots of smoke, flames and molten metal. Therefore each bulb has a special bypass link that fuses and shorts out the bulb if it blows, preventing the arc and keeping the other bulbs lit.

Why might such an awkward arrangement be used? Presumably because it lowers the wiring cost and also because keeps all bulbs at the specified brightness. There is no issue of a bulb at the end of a long wiring run being dim due to voltage drop in the cable. Every lamp will give the specified brightness no matter how far away from the power source. This might be an advantage on an airfield where every lamp must perform exactly to specification.

The wiring cost would be lower because 50 bulbs use the same current as 1 bulb, and therefore you don't need fatter cables to carry 50x the current.

[Brumby]

That's a neat idea.

The point of note is that the load requires a particular current and the supply is made to provide exactly that one specific value of current.  There is no variation in the load current requirement of the lighting system.

The high compliance voltage, however, will require attention to insulation requirements.