21 hours ago, power.esrl3 said:

I found some youtube video that it is fine with TN-C-S. What say you?

Without looking at the video (sorry, time is tight), MEN (Multiple Earthed Neutral) seems to be a type of TN-C-S where there is a neutral to earth bond at the DB. That's what we have in Australia, and multiple bonding is a no-no. So (if I'm right about MEN being a type of TN-C-S), the above statement isn't true for all TN-C-S.

For that subset of TN-C-S where there is no bonding at the DB, then it seems to me that putting a permanent bond at the inverter output would be harmless, though it may be annoying in some way for the network, and may be illegal. I don't know anything about South Africa's regulations.

My notes also say that much of South Africa is TN-S, where there is a bond at the transformer and separate PE and N conductors. In that situation, without thinking about to too hard, it's possibly dangerous to bond neutral to earth anywhere, as you might end up carrying a tiny fraction of neutral current, which could overload your local system. I think in that situation you have to bond the inverter neutral to Eskom's neutral, and trust that connection all the way back to the transformer. But TN-S is based on that trust all the time, it seems to me.

In summary, I don't know enough about the SA situation to be useful; I'm just raising more questions. Sorry.

It seems that this subject would be good to sort out once and for all and publish on a forum like this one. Presumably, (most of) the electrician installers are aware of the situation, but there seem to be enough DIYers who just want to add an inverter for a granny flat or shed etc. and don't want to spend a whole lot on it, and it would be good to know how to do it safely.

42 minutes ago, Coulomb said:

Without looking at the video (sorry, time is tight), MEN (Multiple Earthed Neutral) seems to be a type of TN-C-S where there is a neutral to earth bond at the DB. That's what we have in Australia, and multiple bonding is a no-no. So (if I'm right about MEN being a type of TN-C-S), the above statement isn't true for all TN-C-S.

For that subset of TN-C-S where there is no bonding at the DB, then it seems to me that putting a permanent bond at the inverter output would be harmless, though it may be annoying in some way for the network, and may be illegal. I don't know anything about South Africa's regulations.

My notes also say that much of South Africa is TN-S, where there is a bond at the transformer and separate PE and N conductors. In that situation, without thinking about to too hard, it's possibly dangerous to bond neutral to earth anywhere, as you might end up carrying a tiny fraction of neutral current, which could overload your local system. I think in that situation you have to bond the inverter neutral to Eskom's neutral, and trust that connection all the way back to the transformer. But TN-S is based on that trust all the time, it seems to me.

In summary, I don't know enough about the SA situation to be useful; I'm just raising more questions. Sorry.

It seems that this subject would be good to sort out once and for all and publish on a forum like this one. Presumably, (most of) the electrician installers are aware of the situation, but there seem to be enough DIYers who just want to add an inverter for a granny flat or shed etc. and don't want to spend a whole lot on it, and it would be good to know how to do it safely.

Grateful to hear your expertise regarding this matter. Thank you.

So yeah, I got the time to tinker with my setup.

During ongrid, the voltage between neutral to earth is almost 0.

But during offgrid, voltage is 110v between neutral and earth (oh sh*t).

So I simulate bonding neutral to earth using some insulated wire, and it got a little spark, then I measure again earth to neutral voltage and it is absolute 0. However, I measure earth to neutral amperage, there is some 0.03A, what can you say about this?

So yeah, my ac contactor is arriving, Im rewiring this setup to bond during offgrid.

2 hours ago, power.esrl3 said:

However, I measure earth to neutral amperage, there is some 0.03A, what can you say about this?

I suspect one or more of the loads has a leakage and/or capacitance to earth. 30 mA is right on the limit for tripping a typical Residual Current Device. But if you're measuring this with a clamp meter, it will have inaccuracy plus the usual ±1 LSD measurement error. So it may be a lot less than 30 mA. The inverter itself will had a small current as well, with its EMI suppressing capacitors, but those should be a few nanofarads total, so well under 1 mA.

Perhaps measure the earth current at the DB when the inverter is bypassed, or in a bypass mode.

2 hours ago, Coulomb said:

I suspect one or more of the loads has a leakage and/or capacitance to earth. 30 mA is right on the limit for tripping a typical Residual Current Device. But if you're measuring this with a clamp meter, it will have inaccuracy plus the usual ±1 LSD measurement error. So it may be a lot less than 30 mA. The inverter itself will had a small current as well, with its EMI suppressing capacitors, but those should be a few nanofarads total, so well under 1 mA.

Perhaps measure the earth current at the DB when the inverter is bypassed, or in a bypass mode.

I rechecked the voltage during bypass, earth to neutral is around 1.6V, and current is the same, 0.03A. Yes, Im using cheap clamp meter.

Are you zeroing the clamp meter before clamping? I find that my cheap one has a small offset even on the AC amps range. Of course, it's way worse with the DC amps range, that's the nature of Hall effect devices. 

The latest reply didn't notify me. My setup is all set, thanks much for the help.

If I may ask my own question on here rather than opening a new thread.

I have a 5kw sunsynk inveter, and I have also noticed that if I turn off my mains or the cb to my inverter, that my E/N bonding drops out and I end up with +-130v L-E and +-90v E-N

In my area we regularly are having "visitors" in our local substation that are stealing whatever they can, so I was starting to get into a habit of disconnecting the CB to my inverter to try protect in just incase, however this is obviously a big problem because my earth leakage on my critical board from the plugs does not work due to the loss of bonding (tested with my plug tester).

I want to get my electrician in to fit one of the automatic e/n bonding boxes like the below, but I am also now a bit worried about whether this will cause its own problems if for instance I only disconnect my CB to the inverter.

https://segensolar.co.za/product/segen/protection-segen/ac-protection-box/earth-neutral-bridge-box-for-ac-protection-box/

This would leave my DB with the non-critical loads still powered with their own neutral / earth bond from the municipality then on my critical side of the DB I would have my separate neutral from the inverter but with it also having created its own bonding to the same earth as the mains.

Is this dangerous?

Can it somehow cause the earth or neutral to become imbalanced between the main grid and the inverter?

It appears to me that there is a earthing point at my meter box on my boundary wall as there is a lug with a number of earth cables to it (but I can not see a neutral bonded there so I assume our bonding is done upstream at the substation)

Edited July 31, 20223 yr by Cole

1 hour ago, Cole said:

If I may ask my own question on here rather than opening a new thread.

I have a 5kw sunsynk inveter, and I have also noticed that if I turn off my mains or the cb to my inverter, that my E/N bonding drops out and I end up with +-130v L-E and +-90v E-N

In my area we regularly are having "visitors" in our local substation that are stealing whatever they can, so I was starting to get into a habit of disconnecting the CB to my inverter to try protect in just incase, however this is obviously a big problem because my earth leakage on my critical board from the plugs does not work due to the loss of bonding (tested with my plug tester).

I want to get my electrician in to fit one of the automatic e/n bonding boxes like the below, but I am also now a bit worried about whether this will cause its own problems if for instance I only disconnect my CB to the inverter.

https://segensolar.co.za/product/segen/protection-segen/ac-protection-box/earth-neutral-bridge-box-for-ac-protection-box/

This would leave my DB with the non-critical loads still powered with their own neutral / earth bond from the municipality then on my critical side of the DB I would have my separate neutral from the inverter but with it also having created its own bonding to the same earth as the mains.

Is this dangerous?

Can it somehow cause the earth or neutral to become imbalanced between the main grid and the inverter?

It appears to me that there is a earthing point at my meter box on my boundary wall as there is a lug with a number of earth cables to it (but I can not see a neutral bonded there so I assume our bonding is done upstream at the substation)

Normally N bonding only takes place at the mini sub or on a transformer N. Not multiples as testing is then a nightmare.

On 2022/07/31 at 4:12 PM, Scorp007 said:

Normally N bonding only takes place at the mini sub or on a transformer N. Not multiples as testing is then a nightmare.

For TN-S yes.  But a lot of places in the world including SA also use TN-C-S and TN-C.  Either of those have a single "PEN" conductor.  At the transformer they bonded one of the transformer legs to earth to pull it to earth potential.  That conductors is then PEN, a single conductor running to each installation along with the phase conductors.  Then at the installation (ie. on your property somewhere), you would again split out an earth conductor from PEN at which point it becomes earth and neutral.

6 minutes ago, Gnome said:

For TN-S yes.  But a lot of places in the world including SA also use TN-C-S and TN-C.  Either of those have a single "PEN" conductor.  At the transformer they bonded one of the transformer legs to earth to pull it to earth potential.  That conductors is then PEN, a single conductor running to each installation along with the phase conductors.  Then at the installation (ie. on your property somewhere), you would again split out an earth conductor from PEN at which point it becomes earth and neutral.

In South Africa at the main substation the transformers are connected as ( DYN) delta in the incomer and Y on the output. The neutral comes from the center of the Y winding (theoretically if all three phases are balanced the center of the Y winding will be 0 Volts ) and that is connected to ground via a good earthing system. that point becomes the neutral.

 

4 hours ago, Antonio de Sa said:

In South Africa at the main substation the transformers are connected as ( DYN) delta in the incomer and Y on the output. The neutral comes from the center of the Y winding (theoretically if all three phases are balanced the center of the Y winding will be 0 Volts ) and that is connected to ground via a good earthing system. that point becomes the neutral.

 

This is how I have it as well. As soon as you have unbalanced phases you also get current flowing in the neutral due to the N to E voltage rise.

6 hours ago, Antonio de Sa said:

In South Africa at the main substation the transformers are connected as ( DYN) delta in the incomer and Y on the output. The neutral comes from the center of the Y winding (theoretically if all three phases are balanced the center of the Y winding will be 0 Volts ) and that is connected to ground via a good earthing system. that point becomes the neutral.

@Antonio de Sa you had me scratching my head about this config.  Yes the primary is delta cause no consumer requirement on that side. Secondary must be star cause they need to cater for consumer 220vac as well. So in fact that neatral carries the 220 load currents only.  Which, please correct me, should be 3 times thicker than the 3 phase cores, cause the neatral teturn all 220vac phase loads.

Am i correct? 

 

Just now, BritishRacingGreen said:

 

A null message forum lost my message. Ill try again 😒

12 hours ago, BritishRacingGreen said:

 

In single phase circuits the size of the neutral cable must be the same as the size of the live cable

On 2022/09/23 at 1:25 PM, Antonio de Sa said:

In South Africa at the main substation the transformers are connected as ( DYN) delta in the incomer and Y on the output. The neutral comes from the center of the Y winding (theoretically if all three phases are balanced the center of the Y winding will be 0 Volts ) and that is connected to ground via a good earthing system. that point becomes the neutral.

 

Not sure why you bring this up?  What does it have to do with what I said?

Edited September 24, 20223 yr by Gnome

3 minutes ago, Gnome said:

When using a 3 phase transformer.  The outgoing is wye.  The 3 windings all connect at one point, that point is tied to earth and becomes neutral.  But neutral is a man made concept.

But I still not following what part of this contradicts what I said?  I was talking about earthing systems, specifically at the consumer side which clearly has a pole for the neutral.

Not sure what you are trying to say...

Not disputing what you were saying, I was trying to explain how the neutral is created in our DYn distribution transformers for those who don't know.

2 minutes ago, Antonio de Sa said:

Not disputing what you were saying, I was trying to explain how the neutral is created in our DYn distribution transformers for those who don't know.

Ah ok.  I was super confused.

Btw. broken neutral aside, I don't see how neutral can be anything but "0 volts".  I'm not really sure 0v is the right term.  Across the 3 phases the neutral is the common terminal. It really alone does not have a "voltage".  In relation to each phase it has a voltage/potential.

If it is non-zero in relation to your earth wire, then you have a high impedance earth, but again, nothing to do with balancing the loads.

I assume the utility "balances" the load using taps on the transformer per phase to neutral, and simply changes the taps on the legs depending on how their potential deviates in relation to the required 240 volts.  I've not thought about it too much but I'm guessing that phase to phase they should at that point also be relatively within spec.  Or maybe they change taps on the high side since the wire gauge should at least be thinner and thus more cost effective.

23 minutes ago, Gnome said:

Ah ok.  I was super confused.

Btw. broken neutral aside, I don't see how neutral can be anything but "0 volts".  I'm not really sure 0v is the right term.  Across the 3 phases the neutral is the common terminal. It really alone does not have a "voltage".  In relation to each phase it has a voltage/potential.

If it is non-zero in relation to your earth wire, then you have a high impedance earth, but again, nothing to do with balancing the loads.

I assume the utility "balances" the load using taps on the transformer per phase to neutral, and simply changes the taps on the legs depending on how their potential deviates in relation to the required 240 volts.  I've not thought about it too much but I'm guessing that phase to phase they should at that point also be relatively within spec.  Or maybe they change taps on the high side since the wire gauge should at least be thinner and thus more cost effective.

The tap changer is connected on the HV side of the windings, it allows to increase / decrease the magnetic flux thus having an influence on the LV side load, on our distribution transformer the tap changer is done manually hence one needs to switch off the load on the LV side off to do any change, on the big Eskom transformers they are usually (on Load Tap changers) 

 

Edited September 24, 20223 yr by Antonio de Sa

30 minutes ago, Gnome said:

Ah ok.  I was super confused.

Btw. broken neutral aside, I don't see how neutral can be anything but "0 volts".  I'm not really sure 0v is the right term.  Across the 3 phases the neutral is the common terminal. It really alone does not have a "voltage".  In relation to each phase it has a voltage/potential.

If it is non-zero in relation to your earth wire, then you have a high impedance earth, but again, nothing to do with balancing the loads.

I assume the utility "balances" the load using taps on the transformer per phase to neutral, and simply changes the taps on the legs depending on how their potential deviates in relation to the required 240 volts.  I've not thought about it too much but I'm guessing that phase to phase they should at that point also be relatively within spec.  Or maybe they change taps on the high side since the wire gauge should at least be thinner and thus more cost effective.

Although the 3 phase load will be affected by changing the voltage the only monitor for the tap changer is voltage which they get from the voltage transformers fitted to each busbar in the substation. They do however make use of a line drop compensator circuit. All 3 phases taps are always on the same tap. Common is 11,  14 and 17 positions. Normally done on 66kV and higher transformers on the grid as on line as opposed to off load on distribution transformers.

Anti Islanding to ground the neutral when the grid fails or is off. If the Inverter you have doesn't have a contact that gives 220VAC when the grid falls away one can use the normally Closed contact with the live connected. A time delay on relay should be used to allow the inverter to synchronise. Each inverter is different in the time it takes to switch back to the grid after the grid powers up.  

Edited May 20, 20233 yr by TonyH

On 2021/10/23 at 7:48 AM, RWI said:

ELCBs are old type voltage devices and are not generally used now.  The safety protection devices used in domestic installations and other places today are RCD/RCCB and are current devices.  They measure the current difference between the two wires supplying the device and trip if this is more than a certain value typically 30 mA for domestic installations.  They will therefore work even if there is a high impedance between N and E (not good ground) and if there is no connection at all it is not technically unsafe if one wire is connected to ground unintentionally (fully floating system).  A good N to E connection is however required to trip on earth fault unless you have an IT system (very high impedance earth) used in parts of Europe.

The important thing is that if the system is designed for a hard connection between N and E then this is always maintained preferably by passive means and not relying on actuated devices.  

So if there is an imbalance between L and N the RCD will trip irrespective of an E being wired?

4 hours ago, Leonardus said:

So if there is an imbalance between L and N the RCD will trip irrespective of an E being wired?

Yes the operation takes place when there is a difference between live and neutral. Even equipment that are not earthed like most power tools and hair dryers are protected. 

It is interesting that mention is made of voltage operated ELCB that dates back to 60 yrs ago. In my 48 yrs since my 1st training in the electrical field I have never come across one. 

Yes in this time I have come across many houses that were build before E/L were fitted.