TN-CS Combined With TT Question

[OnlQQker]

So I've just watched this and the chap explains all the different Earth arrangements used in the UK. He mentions about how in a TN-CS earth system if the neutral was broken just before it came into your house then this would be very dangerous. So I was just wondering if a TT system combined with a CN-CS would be safer.
Or am I missing something he is saying and barking completely up the wrong tree? (which is always quite possible)

 

[Julie.]

Sorry about responding a bit late to the party, I haven’t been on-line.

Sorry if it wasn’t too clear – I will attach a few pictures to try to explain.



Firstly, if we didn’t use PME or any bonding to the ground, then each case would look like this:

In this case assuming we are say ½ way along, and the earth impedance is say 400 ohm each side of the lamppost then the potential of the ground at the point of the lamppost would be around ½ of the fault voltage – so say 115 – 120V – if we touch the lamppost – which is held at zero then we have a 120V shock hazard!

If we now bond this lamppost to the ground – say by a 100 ohm connection then the fault current would go through the 400 ohm (A) and then both the 400 ohm (B) and 100 ohm in parallel.

This time then the voltage would only be 38V or so – so a much better situation.

Obviously, there are cases where the earth impedance is lower, but then the earth rod resistance would be lower as well, but for close up situations, it’s possible to get cases which aren’t much better than no bond at all

In practice however we would usually have many similar bonds to ground similar to the picture below

In this case, each point, such as the blob in the centre has a resistance to all the earth bonds around it, the closer ones may be 70 ohm, further ones 120 ohm etc – but if we take an average of 100 ohm and there are 10 of them, then the overall is around 10 ohm; however the real advantage is the effective isolation of each point from more distant ones:

In this case each the two example points end up fairly disconnected from each other – a mini near equipotential zone – in this case the equivalent circuit would look like this:

This gives a touch voltage around 5V or so for the same fault and ground resistance as the simple examples above.

Now, as has been pointed out, these aren’t true equipotential zones as any resistance at all would give rise to a voltage drop, but this is the case throughout all of the real electrical systems – even if you use 10mm^2 to bond everything throughout a house, then there will still be some resistance, and therefore some voltage drop in the event of a fault, but the regs generally use “substantially the same potential” as an EP zone/bonding criteria.

The next issue is the loss of a PEN, so with 500W this would represent a load resistance of 100 ohm or so, our situation would be this:

Effectively there would be a potential of 230V on the body of the lamp, however with the 100 ohm earth rod, it looks like this and we would have around 115V touch potential.

Although this is no good – it is better than having no earth rod, not worse.

Obviously all the above is just an oversimplification, but it illustrates that multiple bonds to ground gives both low overall resistance to ground, plus areas that form equipotential type zones - fairly disconnected from the influences of other areas forming similar zones.

 

[LastManOnline]

Sorry about responding a bit late to the party, I haven’t been on-line.

Sorry if it wasn’t too clear – I will attach a few pictures to try to explain.



Firstly, if we didn’t use PME or any bonding to the ground, then each case would look like this:View attachment 61748
In this case assuming we are say ½ way along, and the earth impedance is say 400 ohm each side of the lamppost then the potential of the ground at the point of the lamppost would be around ½ of the fault voltage – so say 115 – 120V – if we touch the lamppost – which is held at zero then we have a 120V shock hazard!

If we now bond this lamppost to the ground – say by a 100 ohm connection then the fault current would go through the 400 ohm (A) and then both the 400 ohm (B) and 100 ohm in parallel.

View attachment 61749
This time then the voltage would only be 38V or so – so a much better situation.

Obviously, there are cases where the earth impedance is lower, but then the earth rod resistance would be lower as well, but for close up situations, it’s possible to get cases which aren’t much better than no bond at all

In practice however we would usually have many similar bonds to ground similar to the picture below

View attachment 61750
In this case, each point, such as the blob in the centre has a resistance to all the earth bonds around it, the closer ones may be 70 ohm, further ones 120 ohm etc – but if we take an average of 100 ohm and there are 10 of them, then the overall is around 10 ohm; however the real advantage is the effective isolation of each point from more distant ones:

View attachment 61751
In this case each the two example points end up fairly disconnected from each other – a mini near equipotential zone – in this case the equivalent circuit would look like this:

View attachment 61752
This gives a touch voltage around 5V or so for the same fault and ground resistance as the simple examples above.

Now, as has been pointed out, these aren’t true equipotential zones as any resistance at all would give rise to a voltage drop, but this is the case throughout all of the real electrical systems – even if you use 10mm^2 to bond everything throughout a house, then there will still be some resistance, and therefore some voltage drop in the event of a fault, but the regs generally use “substantially the same potential” as an EP zone/bonding criteria.

The next issue is the loss of a PEN, so with 500W this would represent a load resistance of 100 ohm or so, our situation would be this:
View attachment 61753
Effectively there would be a potential of 230V on the body of the lamp, however with the 100 ohm earth rod, it looks like this and we would have around 115V touch potential.

View attachment 61754


Although this is no good – it is better than having no earth rod, not worse.

Obviously all the above is just an oversimplification, but it illustrates that multiple bonds to ground gives both low overall resistance to ground, plus areas that form equipotential type zones - fairly disconnected from the influences of other areas forming similar zones.

Am also late back to the party! You put a lot of work into that reply. Appreciated. Have gotten through most of your calculations and they speak for themselves. You have demonstrated your point that it's better to have an earth rod at the base of the streetlamp than not have one. In your illustrations numbers 5 and 6 (dealing with open PEN) you demonstrated that the touch voltage at the lantern would be 115 volts instead of 230 volts.
I struggle a little however with some of your assumptions.
Firstly 115 volts is "better", but is that acceptable ? , safe? If the lamp wattage is reduced, will the touch voltage not rise higher again?
Secondly to describe the above situation as an "equipotential zone" is employing the term in a way I have not seen before. "equi" meaning equal or same. Its sometimes referred to as a "safe zone". The term conveys the idea of safety in people's minds, sometimes defined as an area where electric shock won't be experienced. I appreciate that you are perhaps implying a "more equal zone" but its important to make the distinction because of the mistaken belief (come across it here in ROI frequently) that the neutral can be "tied down to earth" via a 100 ohm rod. Your own calculations demonstrate that there will actually be a significant PD at the rod.
One final point that has me a little confused is the example you gave of the drunk touching the lantern and receiving a shock coming from the ground which flows back to the substation via the street lamp. Is the drunk not standing on "earth" while also touching "earth" (the street lamp is earthed)? In theory there should be no PD there?

 

[Midwest]

I’ve had to put a couple of lids back on lamp posts. Never seen any earth rods. But isn’t the lamp post an earth rod itself?

 

[LastManOnline]

I’ve had to put a couple of lids back on lamp posts. Never seen any earth rods. But isn’t the lamp post an earth rod itself?

It could be depending on how its secured to earth

 

[pc1966]

Is the drunk not standing on "earth" while also touching "earth" (the street lamp is earthed)? In theory there should be no PD there?

The region very close to the rod (how close depends on the depth of the rod, etc) will be pulled closer to the lamppost, so they should get a much smaller shock than the volts to true Earth.

Again it is not a magic solution, and there will be dangerous voltages possible, but it is definitely less dangerous with the rods!

 

[OnlQQker]

The region very close to the rod (how close depends on the depth of the rod, etc) will be pulled closer to the lamppost, so they should get a much smaller shock than the volts to true Earth.

Again it is not a magic solution, and there will be dangerous voltages possible, but it is definitely less dangerous with the rods!

So electric cars for example could this theory still work?

I'm on about the other end of the supply lead, but I'm not to sure how cars are grounded, though I have seen anti static flaps touching the road from them in the past.

 

[LastManOnline]

The region very close to the rod (how close depends on the depth of the rod, etc) will be pulled closer to the lamppost, so they should get a much smaller shock than the volts to true Earth.

Again it is not a magic solution, and there will be dangerous voltages possible, but it is definitely less dangerous with the rods!

Yes less dangerous. But 115 volts is still VERY dangerous. We seem to have adopted a "river with 2 crocodiles is safer than a river with 4 in it" type of reasoning to some degree here.
Remember this discussion started with the viewpoint that "an earth rod ties the lamppost to SAME potential as the local grounds". Julie, s own calculations disproved that.

 

[LastManOnline]

So electric cars for example could this theory still work?

I'm on about the other end of the supply lead, but I'm not to sure how cars are grounded, though I have seen anti static flaps touching the road from them in the past.

You will find the link "pretty mouth" posted very helpful regarding the use of earth rods with EV chargers.

 

[Julie.]

The point i was trying to illustrate, was that if there is no reliable connection to ground, then there would be a shock risk.

if there is a bond, whether the connection is made using a rod, or incidental via the foundations then that's the situation described where I include a ground connection.

There are some situations where the street furniture is isolated, in some cases the foundations comprise aggregate of stone and sand, which does not make a good connection to ground, and so on.

 

[LastManOnline]

The region very close to the rod (how close depends on the depth of the rod, etc) will be pulled closer to the lamppost, so they should get a much smaller shock than the volts to true Earth.

Again it is not a magic solution, and there will be dangerous voltages possible, but it is definitely less dangerous with the rods!

Have you looked again at the scenario as originally described.? The drunk is standing on the ground when there is a rise in potential at his feet. Now in reality the earths potential is constantly rising and falling with respect to the live. But the lantern is also connected to true earth and hence rises with the ground. So I, m trying to establish how he is at risk of shock. Ironically we have been discussing quite a bit about equiopotential zones. Well true earth is connected directly to the neutral at the substation and the neutral is connected directly to the lanterns earth (TNC-S) so it should not be possible for the drunk to have a PD developing across his body (theoretically)

 

[Julie.]

Yes less dangerous. But 115 volts is still VERY dangerous. We seem to have adopted a "river with 2 crocodiles is safer than a river with 4 in it" type of reasoning to some degree here.
Remember this discussion started with the viewpoint that "an earth rod ties the lamppost to SAME potential as the local grounds". Julie, s own calculations disproved that.

No, they do show that multiple earthing rods do bring the general situation to one of minimal potential difference.

The one particular time it doesn't is when there is an open PEN

The only options to solve this are either to not use a combined conductor at all only TNS, or to ensure that every single piece of metalwork likely to come in contact with line, neutral, or earth(cpc) must be bonded to earth with overall resistances that are approaching zero.

If we take your view that an equipotential zone must have absolutely no voltage differences, then it is purely a theoretical concept as no matter how solid we interconnect anything, in practice there will be some resistance.

 

[Julie.]

Have you looked again at the scenario as originally described.? The drunk is standing on the ground when there is a rise in potential at his feet. Now in reality the earths potential is constantly rising and falling with respect to the live. But the lantern is also connected to true earth and hence rises with the ground. So I, m trying to establish how he is at risk of shock. Ironically we have been discussing quite a bit about equiopotential zones. Well true earth is connected directly to the neutral at the substation and the neutral is connected directly to the lanterns earth (TNC-S) so it should not be possible for the drunk to have a PD developing across his body (theoretically)

No the lamp is not connected to "true earth" as you put it.

That's the whole point of the two scenarios, one has a connection to ground, and one doesn't, by definition the one which isn't connected by a reliable connection will not be connected.

 

[LastManOnline]

No, they do show that multiple earthing rods do bring the general situation to one of minimal potential difference.

The one particular time it doesn't is when there is an open PEN

The only options to solve this are either to not use a combined conductor at all only TNS, or to ensure that every single piece of metalwork likely to come in contact with line, neutral, or earth(cpc) must be bonded to earth with overall resistances that are approaching zero.

If we take your view that an equipotential zone must have absolutely no voltage differences, then it is purely a theoretical concept as no matter how solid we interconnect anything, in practice there will be some resistance.

Agreed. I am comepletely on the same page as you here. Either multiple earthing rods or better still TNS. I was, nt implying that the equipotential zone would be anything different that what we currently enjoy in a properly bonded house.

 

[LastManOnline]

No the lamp is not connected to "true earth" as you put it.

That's the whole point of the two scenarios, one has a connection to ground, and one doesn't, by definition the one which isn't connected by a reliable connection will not be connected.

But is, nt the lantern connected to true earth via the substation?

 

[Julie.]

But is, nt the lantern connected to true earth via the substation?

Which scenario?

With an open PEN, then no! - and that's the situation when there is circ 120V where I said it was better than 240V but no good.

In the other scenarios, yes it's at zero potential, but the ground surrounding it may not be if there is no bonding and there is a fault elsewhere

 

[LastManOnline]

Which scenario?

With an open PEN, then no! - and that's the situation when there is circ 120V where I said it was better than 240V but no good.

In the other scenarios, yes it's at zero potential, but the ground surrounding it may not be if there is no bonding and there is a fault elsewhere

Sorry. I should have described which scenario. So the TNC-S supply system is healthy (no open PEN). The lantern is connected to true earth at the substation and hence connected to the drunks feet. So basically the drunk feets and the lantern are at the same potential. Now let's let take the scenario of the ground at the drunks feet rising in potential with respect to the lantern. You propose that could happen by a line to earth fault at the next lamp. Normally speaking in a healthy TNCS system virtually all the fault current flows back via, the neutral. There will naturally be a small flow down through the DNO, s rod, s depending on the volt drop across the DNO, s neutral and the resistance of the rod. There will also be a "step voltage" extending outwards as there would be with any other rod but that zone would be distancewise a long way from the drunk. So what kind of fault had you in mind that would result in a dangerous PD at the drunks feet?

 

[Julie.]

Sorry. I should have described which scenario. So the TNC-S supply system is healthy (no open PEN). The lantern is connected to true earth at the substation and hence connected to the drunks feet. So basically the drunk feets and the lantern are at the same potential. Now let's let take the scenario of the ground at the drunks feet rising in potential with respect to the lantern. You propose that could happen by a line to earth fault at the next lamp. Normally speaking in a healthy TNCS system virtually all the fault current flows back via, the neutral. There will naturally be a small flow down through the DNO, s rod, s depending on the volt drop across the DNO, s neutral and the resistance of the rod. There will also be a "step voltage" extending outwards as there would be with any other rod but that zone would be distancewise a long way from the drunk. So what kind of fault had you in mind that would result in a dangerous PD at the drunks feet?

Please show which picture!

You state that i have the high voltage in my example - which is the situation where there is no rods fitted - no local earthing arrangements.

Then describe how it can't have a voltage because of the local earthing arrangement.

As examples where there could be a fault to the ground

Ohl cable drops
Next door neighbour cuts lawnmower cable exposing line conductor, which sits in puddle
Individual mis-wires a fitting, connecting the line conductor to the enclosure
....
....

I doubt i said a line to earth fault, as that would not describe the situation I raised

 

[LastManOnline]

Please show which picture!

You state that i have the high voltage in my example - which is the situation where there is no rods fitted - no local earthing arrangements.

Then describe how it can't have a voltage because of the local earthing arrangement.

As examples where there could be a fault to the ground

Ohl cable drops
Next door neighbour cuts lawnmower cable exposing line conductor, which sits in puddle
Individual mis-wires a fitting, connecting the line conductor to the enclosure
....
....

I doubt i said a line to earth fault, as that would not describe the situation I raised

No. You specifically gave the example of " a line to ground fault at the next lamp post". See your post no 12.
You speculate that a voltage of 100 to 230 volts could arise at the drunks feet from such a fault, which in reality could not happen for the reasons I outlined in my last post.

 

[Julie.]

No. You specifically gave the example of " a line to ground fault at the next lamp post". See your post no 12.
You speculate that a voltage of 100 to 230 volts could arise at the drunks feet from such a fault, which in reality could not happen for the reasons I outlined in my last post.

Exactly!


If there is a line to ground fault, the current will pass through the ground, as there is a finite resistance, there would be a voltage drop along the route - perhaps 228V at the point of the fault circ 115/120V 1/2 way along etc.

So if you are stood on the ground at the half way point, you have circ 120V at your feet.

Now if the lamppost is correctly earthed with a solid PEN/TNS type connection this will be held to the Zero potential of the substation. therefore you will be at 120V or so whilst the lamppost is at 0V (or near to).


If you have a fault to earth, then the fault would pass through the PEN/TNS type connection.

 

[LastManOnline]

Exactly!


If there is a line to ground fault, the current will pass through the ground, as there is a finite resistance, there would be a voltage drop along the route - perhaps 228V at the point of the fault circ 115/120V 1/2 way along etc.

So if you are stood on the ground at the half way point, you have circ 120V at your feet.

Now if the lamppost is correctly earthed with a solid PEN/TNS type connection this will be held to the Zero potential of the substation. therefore you will be at 120V or so whilst the lamppost is at 0V (or near to).


If you have a fault to earth, then the fault would pass through the PEN/TNS type connection.

No. That is not correct. Remember we are dealing with a TNC-S system here, not TT. The fault current thus flows back through the DNO, s neutral (largely) and not down through earth. Assuming there is a PME at the lamp post (100 ohms) where the fault occurs than that is simply an extra circuit through which current will flow. But the entire voltage present at the DNO, s rod will be dropped across it just like it would with any other earth electrode here. So not only will there be no dangerous voltage at the "halfway point",(an unspecified distance) there will in fact be no dangerous voltage 10 feet from the electrode as it will have been dropped in its entirety across the (presumed) 100 ohm resistance of the rod

 

[Julie.]

No. That is not correct. Remember we are dealing with a TNC-S system here, not TT. The fault current thus flows back through the DNO, s neutral (largely) and not down through earth. Assuming there is a PME at the lamp post (100 ohms) where the fault occurs than that is simply an extra circuit through which current will flow. But the entire voltage present at the DNO, s rod will be dropped across it just like it would with any other earth electrode here. So not only will there be no dangerous voltage at the "halfway point",(an unspecified distance) there will in fact be no dangerous voltage 10 feet from the electrode as it will have been dropped in its entirety across the (presumed) 100 ohm resistance of the rod

Well if you are talking about when there IS multiple rods etc etc, then of course it ties the local ground to the substation earth.

That's EXACTLY what the calculations I provided shows.

But the description i gave where the voltage distributes along the ground is the situation where there is no multiple earthing - it illustrates the point!!!

I showed the example of large voltage differences due to not having multiple earthing - a potentially large voltage difference if you don't have the earthing in place, compared to low voltage in the event of multiple earthing in place.

 

[LastManOnline]

Well if you are talking about when there IS multiple rods etc etc, then of course it ties the local ground to the substation earth.

That's EXACTLY what the calculations I provided shows.

But the description i gave where the voltage distributes along the ground is the situation where there is no multiple earthing -

Even without multiple earthing the example you gave of a progressive volt drop occurring across the earth between the lamp fault and the substation cannot happen because "true earth" is at or near zero potential. Where the volt drop occurs is at the earth electrode (100 ohms). Let's assume its 150 volts. Well as I said earlier, 10 feet from the rod the ground will be at zero potential and will remain at zero potential through the entire distance back to the substation. If you view the line to earth fault and draw it out as, a, circuit , it's consists of two resistors, the earth electrode (100 ohms) and "true earth" (zero ohms). Ohms laws tell us where and what the voltage will be
Unless the drunk is standing within the "step zone" (say between 1 ato 2 feet of the earth electrode) he will be at no risk of shock. he will be completely safe standing everywhere else

 

[LastManOnline]

I think it's your calculation "where the voltage distributes along the ground" needs to be revisited. The voltage dropping along "true earth" does not distribute evenly like it would when dropping along the DNO, s neutral.
You have assumed otherwise.