Let’s talk Earthing

Looking for advise / regulations around earthing in a solar installation.

Currently what I have is 6mm copper wires linking the panels and structure of each string. They come down to a earth junction outside the outside. From the junction, have 16mm cable going to an earth spike. From the earth junction, also have 16mm cable going towards the 4 combiner boxes. Each combiner box has fuses, surge arrestor and dc isolator. Earth from the surge arrestor joins the 16mm cable going to the outside.

Is the above correct?

For the inverters, I have used the municipal earth.

Inverter has a earth neutral bond and this earth is also just connecting to the municipal earth. I have seen some posts in other Forums where people say this is not correct as when in islanding mode, this earth is also broken. Doesn’t really make sense to me as earth is not really broken even during power failures.

Lastly, I originally had the battery earth also connected to the municipal earth but I have seen some discussion to say that this should also be connected to the earth spike as it’s dc. My concern is that this puts my batteries at risk from lightning induced damage… not talking about a direct hit here. If I create a seperate earth spike for the batteries, then I also open the can of worms for create potential looks which could also cause unnecessary damage from lightning induced surges.

So what’s the correct way to earth inverters, batteries and panels.

Thanks.

I attented the talk by SAGE( N_XX) this week on PV sizing cables and the SANS coming. And the training guy from Potchestroom, forgot his name, said that the PV has to be earthed to the inverter from a earth on the panels.That was not something I expected to hear. I wonder if we could ask @plonkster / @JacoDeJongh advice on how they approach this in the installations. Another thing this guy said who is apparently on the SAPVIA advisory panal said, You dont need to put the cable from PV panels to inverter in metal shroud / Channels, if they are less than 50 meters from the inverter.

Hmm. That’s interesting as the draft sans regulations did mention this. It did speak about 50m limit fir 1 scenario. But in any other section, there was just no limitation of 50m mentioned.

Anyway, just to be safe, I re ran all my string cables in metal sprag. It wasn’t too expensive considering what I’ve spent on everything else and it’s sort of piece of mind in case the regulations change in the future.

1 Like

OK, I am not going to talk about regulations.

I will try and help you understand about the principles of earthing. From that, you can draw conclusions as opposed to blindly following regulations.

What is an earth? Well, I think you have to start with asking what is the earth, in electrical terms. The earth is a great big electrical conductor that is and covers the planet. It is an electrical conductor that stretches from your house in SA to my house in Ireland and it cannot be open-circuited if there is contact with the planet’s surface.
How good is that conductor, well that depends on how you measure it.
It follows the typical proportionalities that are applicable to resistance.
Where resistance is proportional to a resistivity index and the length of the conductor and inversely proportional to the cross-section of the conductor.
What does this mean in practical terms?
Well, dry rock has a high resistivity value and if we were measuring say between two points 10 km apart then you’d expect a high resistance Yes?
Well, not really, because what the planet has going for it is the vast cross-section of its conductor. So over a 10km distance, the resistance between two points can be relatively low.

OK at this point, because we’re big boys I am going to stop using the term “resistance” and start using the term “impedance”. Impedance is an inclusive term that encompasses resistance, capacitance and inductance.

So to get back my statement of “relatively low”, I’d like to give some context. When you see a transmission voltage overhead powerline and its hefty conductors going from town X to town Y, those conductors likely have around a 10%ish difference in impedance than a measurement of the impedance through the ground between town X and town Y.

It can be a 10% lesser impedance than those hefty aluminium conductors. Usually, about 10% more cross-country in a place like SA, especially in the dry season. In a built-up area with pipelines and railway lines or in the wet season it can be 10% ish less. This is a general rule of thumb, just to give readers of this a handle on the numbers.

And that’s where I want to leave this for the time being, because there is a lot to this.
What I want you to take from this is there is a fairly low impedance conductor available at every point on the planet that connects to every other point on the planet.

5 Likes

Next installment.
Note I realize everything I am saying has caveats and exceptions. I am deliberately excluding those things to stay on message, if I was to do otherwise everyone would get lost in the fog.
So, Yes I am dumbing things down, but I think it’s important that I do.

So we have established the earth is a conductor that links all places on the planet. The next very obvious point is human beings are standing on that conductor. And for the most part, we feel safe and are safe from being electrocuted.
We are like birds sitting on a live wire.
Why do those birds not get electrocuted? Well because they are not between two sufficiently different energy points.
If you put your fingers across the positive and negative of an AA 1.5V battery you don’t feel a shock. That isn’t because you aren’t passing electricity, you are. It’s just that you are conducting too little to feel. However, if you lined one hundred AA batteries in series and did the same thing you would feel a shock.
When a bird sits on a live wire, it doesn’t feel a shock. That is because the voltage drop between the one-inch difference between its feet is too small.
If a bird’s feet were further apart it would be electrocuted.
Now, I am just being ridiculous right?
The bird’s feet would have to be hundred’s of metres apart for it to even feel something.
Wrong.
Again it’s useful to get a handle on real-world numbers.
In substations, the earth mat is constructed and the ground surface is gravelled to attain an acceptable " step resistance". That means a sufficiently low enough impedance that a human being should survive the shock when taking a standard stride when the ground beneath him/her conducts electricity.
A single stride, less than a metre.
Now, I must be telling campfire stories.
I saw this live on TV 10 years ago, I was shouting at the TV telling the commentators what was happening. It took the authorities nearly a week to work it out. Until then, everything from black magic to mafia poisonings was mooted.
Read up on it yourselves, there are many accounts:
http://news.bbc.co.uk/sport2/hi/other_sports/horse_racing/9395743.stm
Basically, it was a racecourse parade ring, where the racehorses were being electrocuted because of their long bodies and nice metal shoes making contact with the grass and keeling over dead. Whilst their handlers walking beside them with much shorter strides and higher resistance footwear couldn’t feel a thing.
This happened on the grass on live TV.
The point of this story is to demonstrate that the earth is not only a conductor that human beings are in constant contact with, but it is also a very active conductor.
And although it has a relatively low impedance if the current is sufficient, the volt drop over a very short distance can be lethal.

Those racehorses wouldn’t have died if all their horseshoes had been connected together with a wire.
Current wouldn’t have flowed through their bodies but through the wire, which has a far lower impedance would have only developed a harmless voltage drop.

And so we get to one of the first safety aspects of earthing: Have everything conductored to the same electrical earth point. Do NOT have multiple earth potentials that you as a human can reach, and do not put your fancy electronics on different earths.

Pick an earth, your system earth, an earth spike if you are off-grid, your borehole casing, whatever, but make sure everything that should be earthed is connected to that single equipotential point.

Not doing this and not understanding why it is important, is the single biggest error I see people make when doing their earthing. More earth points, (as opposed to more things earthed) is actually dangerous, I can’t stress this enough.

More tomorrow.

8 Likes

I’m floored. Gobsmacked. Bowled over. Fell off a horse. Have I now heard everything … obviously not. :man_facepalming:

@Phil.g00 This is why I want to belong to a Forum. Where would else would you get a full reply like this?

1 Like

Thanks @Phil.g00

It’s not the answer I was expecting but definitely the answer I appreciate. You are right, it’s better to understand the theory behind it instead of just following the regulations blindly.

That’s a good point and I’m sure this will help others better understand the principals as well.

Okay. I guess this answers my question about the batteries having their own earth.

I think this is a very good point and 1 that most people don’t really understand.

Part 3:
Electricity flows in a circuit. We all know that, yet many of us throw that knowledge out of the window when we consider earthing.
How many people think current flows into the ground and that’s it, it stops?
No, current or charge equalizes with where it came from. It goes back to the source.
I walk across my pile shag-pile carpet in my rubber-soled shoes in a Highveld winter reach for the doorknob and --zap, I feel upwards of a very short 5kV+ shock.
That is still a circuit, I gradually gathered a charge from the earth and zap I delivered it back to the source — the circuit is complete.
So I want you to keep this in mind, current in the ground is returning to its source and if you or your electronics are part of that path there will be a consequence.

OK, back to our AA battery, it has 1.5V between positive and negative. Now if I theoretically rub the negative pole on that same shagpile carpet and it also gains a 5kV electrostatic charge.
This doesn’t mean I have created a 5kV battery, it is still only 1.5V between positive and negative. But it is 5kV above its source which is the earth. So yes, it will still discharge with a nice zap at the same doorknob.

That tiny spark or even a spark that is much smaller that you don’t even feel happens all the time and is enough to destroy sensitive electronics. You will see that people handle sensitive electronics with an earthed wrist strap on a special earthed conductive mat for this very reason. The act of an unearthed hand touching some more sensitive PCB’s is enough to destroy them, way before you feel any static spark.

Anyway, I digress.
Let us look at 4 ways that something that is unearthed can shock a human.
Firstly, I suppose we have to realize that it doesn’t take much current to shock a human. By that, I mean 10’s of milliamps, not hundreds of milliamps is plenty. Secondly, a human is probably connected to earth somehow.

  1. So, the way most people would immediately recognize, is that that “something” is inadvertently in direct contact with a legitimately live conductor. I call that “something” a piece of unearthed metal from now on, ( but it doesn’t have to be).
  2. It picks up an electrostatic charge, (which is like being a capacitor) as in our carpet example.
  3. It acquires an induced voltage due to being in close proximity to a net fluctuating magnetic field. This is is when it acts as the secondary winding of a transformer, and it acts as an inductor.
  4. Everyone’s favourite deserves its own discussion: Lightning.

This brings us to the why do we earth, and what should we earth and will it really affect me questions?

Well case 1, in inadvertently livening an unearthed piece of metal, I think that’s obvious. It will sit there waiting to deliver a nasty shock to the first person standing on earth that innocently touches it. It should be earthed so that the earth current would flow and have tripped the earth-leakage protection before anyone got shocked.

Obvious, isn’t it. Most appliances are already equipped with this earth through a 3 pin plug.
Now, how many of you cut off a three-pin plug and replaced it with a two-pin, or plugged that appliance into a two-pin extension cord without a second thought?

Side note: While I am on this earth-leakage topic and this is a solar forum. There are different types of earth leakages specifically for solar installations. Why? Well, E/L basically devices compare current that goes in with current that comes back. If there is a difference, the E/L says there is a hole in the bucket somewhere and trips.
The thing is conventional legacy devices were made to work with AC. So they could use conventional transformer-based technology. These days certain PV inverters ( transformerless inverters, their name, not mine), still can have an electrical connection through to the solar DC side. This means that a DC earth fault is now a possibility on the load side. So these special E/L units are equipped to be able to measure AC and DC imbalances and then trip accordingly.
A conventional E/L is an AC only device and could be defeated by a DC earth fault.
I have seen these special E/L put forward as special cases to deal with nuisance trippings. To my mind this is misinformation, and if you have the applicable type of inverter you should use the appropriate E/L device. (I am not sure if the regulations have caught up with this distinction yet, and I don’t think installers realize this).
This new device will still function as before in a mains only AC circuit.

Case 2 is a little harder to visualize. Let’s just say a carpet example can scale up massively. I have a background in transmission work. Considering power line maintenance, a line that is disconnected from the power grid, but unearthed is a great big hunk of metal dangling in the breeze. That breeze is the equivalent of shoes on a carpet, rubbing off electrons, creating a charge relative to earth. The electrostatic discharge contains a great deal more energy, and will and has killed power-line maintenance team personnel.
OK. but that power-line was 10kM long with a massive surface area that acts as a huge capacitor plate, I grant you that.
But now start working out the area of that tin roof on your house in the same breeze. That’s also a huge capacitor plate.
Slap some PV panels 100mm above that plate now you have two huge plates pretty close to one another. Plenty of room for a nasty shock, combined with a 3m fall there.
Connect things like this to earth so they constantly discharge and will not build up a substantial electro-static charge. This is the basis for earthing regulations that deal with those details.

Case 3, induced voltage is probably the least likely to affect the domestic setting. Of course, it can kill in an industrial setting.
Probably, the most likely chance of getting a nasty shock at home is if someone livens a coil or winding with a DC source. As it is disconnected, the collapsing magnetic field causes a back emf that can be nasty.

*** One thing I have to add is that an earth is not transferred across a transformer with independent windings. That’s probably all you need to know as a domestic user, put an earth on all windings.
For cabling, you want your AC current to go and come back in the same cable to cancel out its own magnetic field. The net effect can become a problem in asymmetrical circuits involving large currents.
This segways nicely into the next part of the discussion: Case 4 Lightning.

3 Likes

I’m finding this fascinating. Quick question as we get to lightning:
I’ve recently installed a louvre roof over my patio. It is a completely aluminium construction, relatively large (36m2) and stands on top of aluminium poles bolted to the patio floor and the house’s walls.
I live in Cape Town, so lightning isn’t a big deal. However, should this structure be earthed for safety?

OK, I am sorry… but this…

There is also a Tiktok doing the rounds with someone shouting “Ha ha ha da ka ke hadeda…” :slight_smile:

1 Like

I’ve seen that. I couldn’t help but smile. There’s few things in life that upsets me like a clutch of hadidas or swarm of ants.

Lightning is probably the single event that exposes your earthing issues.
I think one of the biggest misunderstandings about lightning damage to equipment is that it has to get hit by a bolt. For sure a bolt of lightning hitting your installation will damage it. But I think a lot of lightning damage is indirect and can be avoided by understanding lightning better.

Let’s consider what lightning is, and maybe some lesser-known characteristics.
Google tells me that there are +/- 100 lightning strikes a second on planet earth, all day every day. I have no idea if these numbers are correct, or if that includes cloud to cloud bolts.

We’ll mainly deal with strikes down to earth and let’s just say there is a lot of electrical current moving around. Actually, it’s twice as much as the strikes, because the discharges from up there have to be fueled by a charge transfer from down here.

So a bolt of lightning is the equivalent of a pinhole breach in massive capacitor plates. Air is a reasonable insulator, but the heat generated by an electrical arc ionises the air and it becomes very conductive. The arc will then sustain itself until the sky charge has dissipated.
Google tells me that a strike contains 1 billion joules of energy, and again I don’t know if that’s true, but let’s just settle on the scientific unit of ten times a helluva lot.

As a bolt hits the earth its charge is trying to match itself with an equal and opposite charge to reach a lower energy state. That means the current has to travel to/from the area of the other capacitor plate (the earth) and meet somewhere along the way.
So picture the strike point current vast amounts of current are flowing outwards from that point in a half-hemisphere shape. Let’s remember that resistance is inversely proportional to the area of a conductor and at the strike point that half hemisphere is very small. As you get further away from the strike point the area increases rapidly and therefore the resistance decreases rapidly.
image

This picture is nice as a demonstration, but it is factually wrong it assumes a direct proportionality of volt drop to distance. When in fact the area of a sphere ( and therefore the conductor area) will increase at the distance squared.
Anyway, my point is according to ohms law the voltage drop in the ground close to the strike point will be massive. Enough to breach any practical insulation you may care to use. It is also a very dangerous place to be standing close to.

So what can we do if we can’t stop it? We do the opposite, we help it on its way.
Current wants to get from A to B, and it’ll go through you or any of your equipment to get there if that’s the easiest path. So let’s give it an even easier path and no reason to go anywhere else.

So bear this in mind current wants to travel in the ground. This is why multiple earths that aren’t equipotential are so dangerous. The charge will come up and earth and down another, and if there is an impedance between them there will be a volt drop and power dissipation. That’s what destroys your equipment.

I am inevitably going to be asked about lightning rods. Later, I think you should be able to answer these questions yourself, but at this point, I just want to say they tempt lightning, but they have their place in certain settings, ( They create an umbrella of direct strike protection over thatched roofs for example).
Something else I better slip in is charge concentrates on sharp points, so metal taller than the surroundings and sharp points tempt lightning.

OK, at this point you should be able to conclude that a small contact area to dissipate a bolt is not as safe as a wider area of earth and the lower the resistance the earth is the better. So if you are using earth spikes, using multiple connected earth spikes over an area. Use them in an area that is naturally damp if you can, and the addition of salt can help increase the conductivity of the ground.
We want lightning to get in the ground as directly as possible, we realize it wants to travel but we want it not to go through our equipment. What are its probable routes? Aquifers, cables, railway tracks and your fence.
How many people think their borehole pump got hit by lightning? A borehole pump is a natural weak point that allows entry onto your copper wiring to travel to another earth. Realize this, provide an alternate path use a robust earth cable back to your main earth.
Your fence can be used to your advantage, in a domestic setting it is in effect a large conductor surrounding your property that should encircle it with a ring of equipotentiality.
On the other hand, a fence that is not earthed back to your main earth will be a source of headaches, it is a second earth point, with all those associated problems. Ever wonder why your electric gate or your security camera keeps blowing in thunderstorms?
For any farmers, with really long fences be aware that you can also import current unnecessarily. If it’s out in the field put some form of an insulated break in your remote fences now and again. Be aware of how dangerous gates are in long fences. That is the one place a human is likely to be, electrically bridge one side of a gate to the other.

My next point is also important, albeit a bit technical. Lightning is essentially a DC phenomenon. We have established that once there is an insulation breakdown, it’s like a hole in a dam, the current flow will just make it bigger. We want the current to use the route we choose first.
But lightning also has an AC component, I am specifically referring to the rise time of the current pulse. It’s very fast, but it isn’t instantaneous. This portion approximates a very high-frequency AC and we can use that to our advantage or not acknowledge it and create problems.
The impedance of an inductor is proportional to frequency, and at the initial rise time of the strike that frequency is very, very high so anything that presents any form of inductance at all is not going present as a good current path.
What can present as an inductor? Well, a coil of wire will be an inductor. So the earth cables that you want to use to sink the current straight to the ground should be devoid of extra coils or bends.
Remember, in the previous discussion, I said induced voltages result from an asymmetrical current. Well, a lightning discharge is asymmetrical, which means:
1.) It can induce damaging voltages in your other wiring nearby by its magnetic effect.
2.) Closeby metal or cabling creates an inductance ( read extra impedance) which might mean the lightning doesn’t go down the nice little rabbit hole you’ve prepared for it.

So in addition to having no extra coils and minimum bends, the earth conductors from places that are likely to discharge should avoid other cabling or proximity to metal where possible.
The corollary of this is to use toroidal ferrite cores around your combined + & - PV dc wires coming in from the roof so that becomes a less attractive route.

While I am discussing the rise time of lightning, I believe that convention surge arrestors are too slow to afford any real protection. They will certainly not hurt to have but I recommend using transil diodes as well as the conventional type. These are cheap, can be bi-directional or uni-directional and can be ordered at numerous breakdown voltages. They about a thousand times faster at entering avalanche mode.
They are not necessarily sacrificial either and reseal, but who cares if they save 10K of kit.
A 30kW transil is only about the size of a 2W resistor. I suggest you use 3 transil diodes on your DC busbars, + to earth, - to earth ( unidirectional) & + to - ( bi-directional), rated about 30V higher than your highest working voltage.

I think we’ve covered how to protect yourself as current travels in the earth.
There is something else I haven’t touched on. That is worth bearing in mind.
Before a lightning strike, as the sky builds up a charge, things on the ground underneath do too.
If the cloud is negatively charged, the tall things on the ground will start building up a positive charge. Then the insulation between the sky and earth breakdowns with a strike and the sky charge is neutralized. ( Or perhaps even a cloud to cloud strike).
So, what about those tall things that were charged that didn’t get equalized, the sky is now uncharged and no longer attracting. So those tall things must discharge as well. Not as powerful as lightning, but enough to make a person get a nice tingle when they’re washing the dishes. So you see things can be damaged not just from a direct hit or lightning’s own dissipation in the ground. They can be damaged just by things on your own property discharging theoretically just from a cloud to a cloud strike.
Similar earthing principles apply, so don’t necessarily assume you can’t have issues because you never get struck.

So to conclude, if you use a single electrical earth and it is a good one and if you realize the invisible forces at play, with a bit of common sense you should be able to work out why that troublesome piece of equipment keeps getting fried.
Watch that you don’t create unnecessary inductances in your earth cabling. Realize the roles that things like boreholes, roofs, plumbing and fences play as automatically presenting as second electrical earth.
Observe the regulations, and when in doubt it is better to earth ( to a single earth, I can’t stress that enough), than not.
Don’t try to stop lightning, but try to understand it and guide it.

7 Likes

Glad you mentioned this was having a chat to a electrical engineer about this just last week and it makes total sense. Why put in multiple earth spikes when your fence is already doing this, just need to connect fence to your main earth spike.

1 Like

I have a question: Sometimes I get ideas, most of the time they are stupid, and therefore it is a good idea to the test it with people.

We seem to be using the earth as a conductor, to a large extent, this would include the water bodies, probably the ocean mainly.

Now, presumably water (salt water like the ocean) isn’t a too bad conductor, but it likely still has quite a bit of resistance compared to something like a copper wire?

Given this resistance, doesn’t it heat up if we use it as a conductor? If yes, this raises a further question for me: Since all of our electrical losses end up as heat, how much of it would end up as heat in the ocean (which presumably is a bad thing to heat up)?

Hi @jykenmynie , I am only aware of using the ocean and rivers for conducting electricity in remote/rare and extreme conditions. I am not aware of the large extend. We might in remote cases pass current through a water body, but its not the same as sending current into an element. An element can boil water, but the same current flowing through the wires to that element does not raise the temp of the wires to the same level. If designed correctly the current flow through a cable will only raise the temp marginally.

In the one case I learned about in China in 2012, they had an VSD Driven compression on an “oil rig” like platform in the ocean. It used 2 phase 11kv. They ran 2 separate “Multistrand” but single core cables from the shore to the rig, and one got damaged and ripped off with the ends ending up a distance from each other. The Chinese government at the time did not approve the replacement of the cable and the engineers started looking at options, they ended up securing the 2 ends as close together as they could and they removed the insulation from both ends. They switched on the power and the compressor worked. The salt water was sufficient to act as a conductor and by the time of this meeting with the company it was running like that for a few years already.

Now to come back to heating up the ocean.

Movement of sea water on the other hand is measured in SV and 1 SV = 1000000 sqM (35 000 000 cu feet) of water per second. This means that even if the current flow to this 4 Megawatt 11kv compressor did generate some kind of heat, it will take ages to heat that amount of water taking into consideration that every second millions of litres of “unheated” water can pass that area every second.

Personally, I dont think that the amount of electricity we currently send through water bodies can really play any significant role in heating up the ocean or even parts of the ocean with its estimated 1.333 billion cubic km of water.

Edit: I was interested in the resistance of sea water and found this

From this we should be able to compare heat losses.

1 Like

Except for a few extremes like @JacoDeJongh mentioned and another reference about a return current going via San Francisco bay (which I can’t find now) Santa Monica bay, we’re mostly using earth as a “sensing” conductor and reference these days so that safety devices work.

I know that in the UK they seem to prefer low-resistance paths to earth so that a normal breaker trips when there’s a short to earth, but we seem to prefer higher-resistance paths so that the fault current is limited. It might have to do with the state of electrical installations in the country and what would be safest given what’s already there, but I’m out of my depth here.

Edit: Santa Monica bay ref

Each of the three legs in the feeder carrys 600-800 Amps (depending on demand) of 230KV three phase power. The ground return is the Santa Monica Bay. Down at the Scattergood Steam Plant and up in Santa Monica they have a giant copper anchors out in the bay.

Okay now I am out of my dept, I have no idea what heat could be caused by passing current through water, but looking at this:

I can calculate what the wattage loss would be if it was a copper cable.

Sending 11 kv at 404 amps through a .5mm (23ohm) copper wire over 150 meters would result in a 510Volt (4.64%) volt drop.

Simplified version. 404A x 510W = 206kw (element)

Using this in an online water heating calculator:

Weerstand 3

So if the compressor was running continuously through a 0.5mm sq copper wire it would take 627 h 47min (26 days) to raise the temp of 1 000 000 L still standing water by 1 deg C.

All laws of physics ( i.e electrical laws still apply).
We (and nature) pass incredible size currents through the planet’s crust (land and ocean alike).
Remember though resistance is also a function of the cross-sectional area of a conductor.
R = ((Rho)*Length)/Area
(Where Rho is the specific coefficient of resistance for that material).

So relatively low resistances can be achieved through what may be thought of as a non-conductive material. That is if its cross-sectional area is massive.

Yes, it will heat up according to I2R losses, just like any conductor, but again the massive volume of the conductor comes into play and the heat dissipation is negligible per unit.

Incidentally, the DC lines from Cahorra Basa were sometimes run with a DC pole and the return leg through earth. This was when there was terrorist activity in times gone by and one leg of DC line was sabotaged.

No in ZA, they prefer low-resistance paths to earth, from the power station gen transformer to the wall socket in a domestic house. In fact, a low resistance earth-loop test rest is a COC sign-off requirement.

It seems counter-intuitive to want high fault currents during a fault, I know.

But suffice it to say, in an unearthed/earth-compensated system there are also additional insulation problems to consider, so it is a choice.
(Edit: ZA does have earth fault current limiting systems in some places, but these still allow a fairly substantial earth-fault current, so as not to mask an earth fault. Some countries allow a single earth fault to remain on the system, whilst maintaining supply. ZA doesn’t do that, at least ESKOM doesn’t anyway).

( I can go into this but it is technical).
An unearthed system can run with a single existing earth fault, that won’t damage equipment, but the capacitive leakage current ( which may be less than 10Amps at system voltage), is still sufficient to be lethal. So there are safety concerns around human life.

With a conventionally earthed system an earth fault is not masked, the earth fault current is massive and the system cannot sustain it, so the plant shuts down very fast. And in theory, there is less ongoing risk to human life.

1 Like

I have an old steam iron that uses salt water to make steam, with two carbon electrodes…

Also, In Southern Namibia there are quite a few SWER lines (single wire earth return) where the earth is used as an actual conductor… :slight_smile: