Generator feed to 2L+1N system

[kevindd992002]

If one end of the output is internally grounded (i.e. connected to the frame, ground terminal or receptacle ground prong) then a continuity test will reveal this by a very low resistance (<0.2 ohms) from the neutral / 0v side to that point. You will also see a low (but not as low) resistance from the hot (230V) side via the generator windings. In this case you cannot use an autotransformer to create a grounded centre-tap, because two points at different voltages would then be grounded and hence connected together.

Ok, that makes sense. So basically I need to check if my gen's output is L1=L2=110V for me to be able to use an autotransformer.

An isolating transformer (with separate primary and secondary windings) would be OK as there would no longer be any connection between the generator output and the panel.

Are you saying that if my gen's output is L1=230V and L2=0V (internally grounded), a two-windings transformer will work to get a neutral and solve my 110V problem? Can you explain how as I'm confused by how this would work. If there's a schematic diagram out in the Internet that you know explains this, please link me to it.

If the output is floating you will read no continuity from either side to the frame / terminal / ground prong. This is by far the more common arrangement for mobile / portable generators. Although the two wires of the output might be called 'hot' and 'neutral' and coloured to match, if neither is grounded there is no hot or neutral as there is no defined voltage to ground; they are just two wires forming a circuit.

If you ground one of them, you force it to 0V with respect to ground, so it becomes the neutral, and as the other is 230V from it, and hence now 230V from ground, it becomes the hot. If instead you ground an artificial centre-tap to make that the neutral, you force both output terminals of the generator to 115V with respect to ground, so they both become hots.

Ok, this makes complete sense. I was wondering about the hot, neutral, ground terminologies for floating outputs.

 

[pc1966]

Ok, that makes sense. So basically I need to check if my gen's output is L1=L2=110V for me to be able to use an autotransformer.

Check the resistance, it is safer and more dependable!

Basically if the alternator is floating you might see any combination of voltages from either L/N terminal to E when tested as it depends on the winding capacitance for the return path.

 

[kevindd992002]

As Lucien said - you really need to check the resistance (with generator off and locked from starting!) between the L & N and between either of them and the E / chassis connection.

If L-N shows a few to ten-ish of ohms that is normal and just the alternator winding resistance, and it also tells you the alternator is being tested here. Then go L-E and/or N-E, if both show "infinite" impedance then you can safely use it with a false neutral from an autotransformer.

Otherwise you would need an isolating transformer with a centre-tapped output and they are significantly more expensive.

Check the resistance, it is safer and more dependable!

Basically if the alternator is floating you might see any combination of voltages from either L/N terminal to E when tested as it depends on the winding capacitance for the return path.

Ok, I will test the resistances between the terminals and terminals to internal chassis ground this week and report back.

Example of how I would probably wire in a 220V generator for a 110-0-110V split phase load.

Here the N-E link is in the DB American-style, but in the UK that TN-C form is not permitted as it would have to be at the utility's side TN-C-S form.

Again, you don't want the installation earthing system to ever be disconnected no matter what point in the operation of the change-over switch you are at.

Thanks for the diagram. This is exactly how I expected the whole system to be connected. A picture (in this case a schematic) really paints a thousand words!

 

[kevindd992002]

Ok, so I just checked the resistances and here's what I got:

L1-E -> no continuity
L2-E -> no continuity
L1-L2 -> 0.5 Ohms (is this the resistance between the windings? I was assuming about 10 Ohms or so)

So I'm guessing I have a floating 230V output here and I can use an autotransformer to get that centre-tap I need.

 

[pc1966]

Ok, so I just checked the resistances and here's what I got:

L1-E -> no continuity
L2-E -> no continuity
L1-L2 -> 0.5 Ohms (is this the resistance between the windings? I was assuming about 10 Ohms or so)

It all depends on the generator output. 0.5 ohms is pretty reasonable for a 3kVA sort of unit as the I2R losses are not much more than 1% or so.

So I'm guessing I have a floating 230V output here and I can use an autotransformer to get that centre-tap I need.

Yes, looks fine for that! Again, just check the transformer kVA is reasonable (say 50-100%) relative to the generator's output. You might get away with less than 50% if your 120-0-120V loads are fairly well balanced, but safely protecting it is probably more trouble and cost than getting a bigger one.

 

[kevindd992002]

My generator is a 10.8KVA model though. Is 0.5 ohms still normal?

Yeah, at normal operation I do see just about 0.5A or so of return/neutral current so I'm guessing I can get away with a 50% (5.4KVA) autotransformer.

 

[kevindd992002]

I just checked and the highest autotransformer power rating I see around is 3KVA and it is expensive at around $230.

 

[kevindd992002]

When connecting the generator earth to the earthing system of my house, can I directly just run a grounding wire from the ground output terminal of the generator to the ground bus bar of the main panel? I know it's better to connect them directly to the ground electrodes but I can't do that anymore because they are already buried.

 

[Simon47]

Before going down the route of buying transformers etc, it would be worth looking to see what the generator has in terms of windings. Just gently popping the screws out and seeing what's behind the electrical panel could be instructive.
I'm thinking that because 120V is so common, and 120/240V switchable is so common, there's a good chance that you have 2off 120V windings connected to give 240V only. If that's the case, it would be very simple to connect to the joint connecting the two windows together to get your neutral.

 

[Avo Mk8]

Hi Simon - intrigued by this, I had been trying to find a spec or schematic of the generator in question, but to no avail. Would like to hear from anyone who has access to info.
I believe this is it - 'popping the cover off' may be a slight simplification, but worth a try:

 

[pc1966]

My generator is a 10.8KVA model though. Is 0.5 ohms still normal?

Er, no as 10.8kVA = 45A at 240V so that would be 10% of the power lost in copper windings from I^2R!

Might be measurement error, as a generator with somewhat optimistic short-term ratings...

Yeah, at normal operation I do see just about 0.5A or so of return/neutral current so I'm guessing I can get away with a 50% (5.4KVA) autotransformer.

Yes, in fact you could probably get away with 2kVA normally BUT it is hard to protect the transformer and not risk on overload of it that neutral is disconnected (or goes to one of L1/L2) and causes serious over-volt damage to whatever is imbalanced on your L1-N-L2 arrangement.

 

[Simon47]

popping the cover off' may be a slight simplification, but worth a try:

Six screws andvthe pabel with the voltmeter & sockets should come off. Lets see what's behind it - in particular how many wires come out of the generator itself.

 

[kevindd992002]

We decided to leave out the neutral from the generator for now and just do two things:

1. Include the neutral wire from the post to the transfer switch mechanism so that the neutral line gets cut off from the main panel when the generator is used. No 110V is better than a very unstable 110V that we have now when the generator is working.

2. Include the ground of the generator to the whole earth system.

As for #2, I have some clarification. So in this house there are two main panels, one for each of the two buildings. They both get L1, L2, and N from the same electricity meter. Each of the main panels has two ground rods bonded to the ground bus bar. Only building 1 (B1) is connected with the gen set/transfer switch circuitry as it is the main house. Building 2 (B2) has no power when there is a blackout.

Now the problem is that the generator is very close to B2 and would be very hard to include it to the earth system of B1 as it is farther away from the building. Is it fine to simply connect the metal of the generator to the ground bus bar in the main panel of B2 and call it a day? Or is the proper way really to find a way to connect it back to the ground rods of B1? Is there even a difference since all of these are grounds anyway?

 

[Simon47]

We decided to leave out the neutral from the generator for now and just do two things:

1. Include the neutral wire from the post to the transfer switch mechanism so that the neutral line gets cut off from the main panel when the generator is used. No 110V is better than a very unstable 110V that we have now when the generator is working.

Sorry, but that does NOT mean you don't have 110V, you still have a split system where your 110V circuits will be anything from nothing to the full 220V. Your only guarantee is that the two side of 110V will add up to the 220V from the genny.
It also means that your overcurrent protection may or may not work in the event of a fault - it won't work at all for a L-E fault.

2. Include the ground of the generator to the whole earth system.

As for #2, I have some clarification. So in this house there are two main panels, one for each of the two buildings. They both get L1, L2, and N from the same electricity meter. Each of the main panels has two ground rods bonded to the ground bus bar. Only building 1 (B1) is connected with the gen set/transfer switch circuitry as it is the main house. Building 2 (B2) has no power when there is a blackout.

Now the problem is that the generator is very close to B2 and would be very hard to include it to the earth system of B1 as it is farther away from the building. Is it fine to simply connect the metal of the generator to the ground bus bar in the main panel of B2 and call it a day? Or is the proper way really to find a way to connect it back to the ground rods of B1? Is there even a difference since all of these are grounds anyway?

Are the grounds for B1 and B2 linked ? Linking them would be a good idea anyway as it'll make your combined ground significantly better than either of them on it's own.
But just using the ground from B2 for the genny and the ground for B1 for the loads in B1 is much like the distributors do in TT distribution systems - there they ground the neutral at the transformer and thus ensure that there is a current path for line-earth faults to trip overcurrent protection.

But, this all supposes you have a neutral from the genny - that would be a centre tap to a 220V winding, or the connection between the windings where the 220V output is from two-off 110V windings (which I suspect you might have).
Without that neutral connection, you do NOT have a safe setup. Earthing the frame of the genny will not help.

To be clear on this, your setup is unsafe as you propose.

As I've mentioned before, just undo the six screws holding the panel on with the meter and sockets, and ease it forward very carefully. We're interested in how many wires come out of the generator itself, and how they are connected. You may find your generator has everything you need to provide a safe supply with 110V circuits correctly supplied.

The alternative is to use an auto-transformer to generate a synthetic neutral.
This is not the same as a transformer to convert your 220V into 2x110V and handling the full load. This is a transformer with nothing more than 2off 110V windings (or a single centre tapped 220V winding) which connects across the 220V from the generator - with the centre tap connected to the neutral of your board. If the two sides (L1-N and N-L2) are unbalanced in voltage, then a current will flow in the transformer to balance the system.
The transformer will only handle the difference in 110V currents - e.g. if one side of the board had (say) 10A of loads, while the other had (say) 5A - then the transformer "neutral" would pass 5A in order to balance the loads, which would be split as 2.5A in each half of the winding, with the generator seeing a 7.5A load. Worst case would be only one side of the 110V loaded where the transformer supplied the whole current.

 

[kevindd992002]

Sorry, but that does NOT mean you don't have 110V, you still have a split system where your 110V circuits will be anything from nothing to the full 220V. Your only guarantee is that the two side of 110V will add up to the 220V from the genny.
It also means that your overcurrent protection may or may not work in the event of a fault - it won't work at all for a L-E fault.

Are the grounds for B1 and B2 linked ? Linking them would be a good idea anyway as it'll make your combined ground significantly better than either of them on it's own.
But just using the ground from B2 for the genny and the ground for B1 for the loads in B1 is much like the distributors do in TT distribution systems - there they ground the neutral at the transformer and thus ensure that there is a current path for line-earth faults to trip overcurrent protection.

But, this all supposes you have a neutral from the genny - that would be a centre tap to a 220V winding, or the connection between the windings where the 220V output is from two-off 110V windings (which I suspect you might have).
Without that neutral connection, you do NOT have a safe setup. Earthing the frame of the genny will not help.

To be clear on this, your setup is unsafe as you propose.

As I've mentioned before, just undo the six screws holding the panel on with the meter and sockets, and ease it forward very carefully. We're interested in how many wires come out of the generator itself, and how they are connected. You may find your generator has everything you need to provide a safe supply with 110V circuits correctly supplied.

The alternative is to use an auto-transformer to generate a synthetic neutral.
This is not the same as a transformer to convert your 220V into 2x110V and handling the full load. This is a transformer with nothing more than 2off 110V windings (or a single centre tapped 220V winding) which connects across the 220V from the generator - with the centre tap connected to the neutral of your board. If the two sides (L1-N and N-L2) are unbalanced in voltage, then a current will flow in the transformer to balance the system.
The transformer will only handle the difference in 110V currents - e.g. if one side of the board had (say) 10A of loads, while the other had (say) 5A - then the transformer "neutral" would pass 5A in order to balance the loads, which would be split as 2.5A in each half of the winding, with the generator seeing a 7.5A load. Worst case would be only one side of the 110V loaded where the transformer supplied the whole current.

Ok, so for #1 you lost me there. How can I not "not have a 110V" with that proposed setup if when I transfer to generator power, the L1-N or L2-N of all my 110V outlets do not have a complete circuit? I only get L1 and L2 from the gen and the N terminal of the 110V sockets are practically open. So my 220V sockets will have a complete circuit while the 110V sockets won't at all.

No, the B1 and B2 ground rod pairs are not linked together. We treated them as separate buildings and linking them together was very hard because when the house was renovated, the grounding system implementation came late to the party (don't ask whose fault was it).

You're right, without the neutral from the generator connecting the gen to the B2 ground will not really trigger overcurrent protection because the an L-E fault will return the fault current to the neutral path to the transformer and not to the generator neutral, making it unsafe. But doing so will still protect metal objects connected to the generator when lightning strikes, no? I know that that is the main purpose of using ground rods as well explained here.

I'm curious though, if my gen output has a floating 230V (L1-L2) and ground terminals, how does it provide overcurrent protection (during an L-E fault) to a building in general? It doesn't, right? It only has overload protection to trip the breaker (which I doubt will happen because the gen itself will be the first one to be overloaded).

I understand my choices for the generator to be able to supply a neutral conductor. But then, let's say I do have a centre tap or I'm able to provide one, my problem then would be the same as my ground problem in my previous reply, there is no way to connect the neutral of the gen back to the main panel in B1 without destroying several house structure.

 

[Simon47]

Ok, so for #1 you lost me there. How can I not "not have a 110V" with that proposed setup if when I transfer to generator power, the L1-N or L2-N of all my 110V outlets do not have a complete circuit? I only get L1 and L2 from the gen and the N terminal of the 110V sockets are practically open. So my 220V sockets will have a complete circuit while the 110V sockets won't at all.

OK, take two 110V light bulbs (in holders), connect one wire of each to L1 and L2 respectively, and connect the other wires together. You now have a circuit from L1 through one bulb, through the other bulb, to L2. If the bulbs are the same rating then the midpoint will be half way between L1 and L2 so 110V across each bulb.
Follow me so far ?
No consider if instead of each bulb, you have the loads on the L1 and L2 side of the distribution board. With no neutral connection, power will flow from L1, through some loads to the neutral bar in the DB, then through some loads to the L2 connection. IF the loads are equal then they will all see 110V, but if the loads on the two sides of the boards are different then some will see higher and some lower than 110V - as you've already observed.

You could only "not have any 110V loads" if your DB had separate neutral bars for the two sides of the board AND you (or your electrician) had been very careful to use the right neutral bar for every circuit AND you separately switched the neutrals. And if you achieved that, it would be a dangerous setup because then ALL your 110V circuits would be live - connected to a live supply but with no neutral. In the UK that would not be compliant with the regulations (it's not allowed to put switches in the neutral side of a circuit.)

No, the B1 and B2 ground rod pairs are not linked together. We treated them as separate buildings and linking them together was very hard because when the house was renovated, the grounding system implementation came late to the party (don't ask whose fault was it).

Are you sure there's no way to run a cable between them ? it's only a single core cable, and it doesn't have to follow any specific route (within reason, as long as it's not exposed to undue risk of damage).

I'm curious though, if my gen output has a floating 230V (L1-L2) and ground terminals, how does it provide overcurrent protection (during an L-E fault) to a building in general? It doesn't, right?

Correct, there will be no (line-earth) fault protection. What would happen in the even to (say) an L1-E fault is that yor L1 line would be "earthed" via the fault, and the L2 line would go to 220V relative to earth. It would not be immediately dangerous as there should be nothing touchable that's live - and this is the principle of an IT system. Only if a second fault (L2-E) occurred would anything trip - and then either or both breakers for the two loads would trip.
There will be overload and line-line fault protection as either of these would result in excess current flowing around the circuit and trip with the breaker/fuse in your DB or the breaker on the genny - or possibly just cause the genny output to collapse as many small generators are incapable of producing the current needed to trip larger breakers/fuses.

I understand my choices for the generator to be able to supply a neutral conductor. But then, let's say I do have a centre tap or I'm able to provide one, my problem then would be the same as my ground problem in my previous reply, there is no way to connect the neutral of the gen back to the main panel in B1 without destroying several house structure.

Ah, I see part of the problem - you only have a 2 core (or 2 core plus earth ?) cable from the house to the genny location, and it's not easy to run a new cable ?
I think your best bet is to take 220V to the house and use an auto-transformer to make a synthetic neutral. You'd treat it as an IT supply, and it's possible to configure an RCD to trip the supply (at the house end of the cable) in the event of a L-E fault within the house.

 

[kevindd992002]

OK, take two 110V light bulbs (in holders), connect one wire of each to L1 and L2 respectively, and connect the other wires together. You now have a circuit from L1 through one bulb, through the other bulb, to L2. If the bulbs are the same rating then the midpoint will be half way between L1 and L2 so 110V across each bulb.
Follow me so far ?
No consider if instead of each bulb, you have the loads on the L1 and L2 side of the distribution board. With no neutral connection, power will flow from L1, through some loads to the neutral bar in the DB, then through some loads to the L2 connection. IF the loads are equal then they will all see 110V, but if the loads on the two sides of the boards are different then some will see higher and some lower than 110V - as you've already observed.

Ok, that makes more sense now. Electricity will take all available path to get back to the source and with no neutral connection, current the loads will act like one to complete the circuit for the return current. So that has been my problem all along. I thought it was because the neutral from the power meter is directly connected to my main panel and not included in the transfer switch mechanism. So with the neutral from the power meter directly connected to B1's main panel, it effectively does nothing when B1 is on generator power? Also, why is it that when I measure across the L-N of a 110V socket I see unstable voltage from 110V to around 200V? From you explanation, shouldn't I be seeing stable but varying voltage instead?

You could only "not have any 110V loads" if your DB had separate neutral bars for the two sides of the board AND you (or your electrician) had been very careful to use the right neutral bar for every circuit AND you separately switched the neutrals. And if you achieved that, it would be a dangerous setup because then ALL your 110V circuits would be live - connected to a live supply but with no neutral. In the UK that would not be compliant with the regulations (it's not allowed to put switches in the neutral side of a circuit.)

Is there a use case for this though? If you had separate neutral bars for L1 and L2, how do the 110V circuits be live if they still are connected to the neutral in the DB?

Also, are you saying that in the UK the neutral is always directly connected to the main panel and cannot be part of the transfer switch like what I have in my current setup?

Are you sure there's no way to run a cable between them ? it's only a single core cable, and it doesn't have to follow any specific route (within reason, as long as it's not exposed to undue risk of damage).

Well, I mean I can run a cable between them, sure, but the ground rods are already buried under cement and that would entail destroying that cement floor again. Too much work and probably an expensive one. What is the disadvantage of having them disconnected anyway? Isn't this the same as like two different houses in a TT system that are not owned by the same owner? I thought of it like that which is why I did not have them connected in the first place.

Correct, there will be no (line-earth) fault protection. What would happen in the even to (say) an L1-E fault is that yor L1 line would be "earthed" via the fault, and the L2 line would go to 220V relative to earth. It would not be immediately dangerous as there should be nothing touchable that's live - and this is the principle of an IT system. Only if a second fault (L2-E) occurred would anything trip - and then either or both breakers for the two loads would trip.
There will be overload and line-line fault protection as either of these would result in excess current flowing around the circuit and trip with the breaker/fuse in your DB or the breaker on the genny - or possibly just cause the genny output to collapse as many small generators are incapable of producing the current needed to trip larger breakers/fuses.

I'm only familiar with a TT system as that's what I have, so forgive me, but if there's an L1-E fault how does it not become dangerous if, say, a microwave (with metal exterior) is the one that caused that fault and someone touched that metal? Are you saying that the microwave simply won't work when this L1-E fault happens and wouldn't affect the other loads because the voltage between L1 and L2 will still be technically 220V (L2 being 220V relative to the faulted L1)?

Ah, I see part of the problem - you only have a 2 core (or 2 core plus earth ?) cable from the house to the genny location, and it's not easy to run a new cable ?
I think your best bet is to take 220V to the house and use an auto-transformer to make a synthetic neutral. You'd treat it as an IT supply, and it's possible to configure an RCD to trip the supply (at the house end of the cable) in the event of a L-E fault within the house.

Correct, I only have a 2 core cable running from the house to the genset location and it's definitely hard to run a new cable for both neutral and ground without exposing them.

How do I treat the genset source as an IT supply when an auto-transformer is used? I thought an IT system doesn't have a centre-tapped neutral that the auto-transformer provides? For RCD, aren't they better to be installed on the different subpanels instead of having them in the main panel and just use them on breakers that supply places like bathroom (or any places that can be wet) where a person can potentially touch a live wire and be dangerously part of the circuit?

 

[Megawatt]

I live in the Philippines but I have the same US electrical system (2 x 110V hot + 1 x neutral). I have two ground rods buried 8 ft. deep and connected to the main panel. Netural and ground are bonded in the main panel. And from the main panel, all 4 wires (2L + 1N + 1G) go to sub panels. Ground and neutral are isolated from each other in the sub panels as they should be. From the sub panels, I can create 220V sockets (2L + 1G) and 110V sockets (1L + 1N + 1G).

A generator is also installed in the same house but the generator only has three output terminals (2 x 110V hot + ground). Only the two 110V terminals are fed to the transfer switch. When the power goes out and the transfer switch is switched to the generator feed, I don't get grounding to my overall system and I also don't get neutral. The 220V sockets work just fine ( though without ground) but the 110V sockets are funky since I don't have any neutral feed from the generator. Their output voltages fluctuate from 130V-150Vac when using a volt meter. Is this to be expected? Was a wrong type of generator installed?

Kevin what is the wattage on your generator ? I have never seen a generator that didn’t have a neutral and also it should have a 30 amp 4 wire twist lock receptacle also. I’m assuming it is a very low wattage. All generators have a neutral and I would guess that your voltage regulator is probably bad. You would be better off if you just bought a new generator with at least 5000 watts I have one and it’s 10,000 watts. Do not continue using your generator with no neutral. With out a neutral Your voltage is going to fluctuate and possibly cause damage to your appliances which operate at 120vac like your TV, microwave etc. the neutral in the generator is supposed to be bonded to the frame. Good luck

 

[Lucien Nunes]

@Megawatt this is not a US-spec generator, it's European spec, so it has only two wires - a neutral and a 240V hot - plus a frame ground. (We don't have 4-wire circuits or twistlocks!) The problem is that the OP has US-spec electrical wiring with a mixture of 120 and 240V appliances and panels with two 120V hots and a neutral. To run the 120V appliances and lights from the generator requires a transformer, or if the generator output winding has a hidden centre-tap, modifying its outlet box to bring that out as the neutral instead and make the existing 240V neutral into a second 120V hot.

As yet we haven't seen inside the generator we don't know whether the tap in the winding exists (it does in some units so they can be sold globally by fitting the correct outlet box). In any case it sounds like the cable to the generator location does not have the 3rd wire to allow for that modification and is difficult to replace. So the best solution so far is to send the generator's 2-wire 240V to the house with neither wire grounded (I.e. make the neutral a hot by not bonding it) and there use a transformer to create a neutral point halfway between the two, and bond that instead. A 3-pole GFCI on the transformer with the bonding connection to the transformer side of the GFCI will provide shock protection both at the generator and in the house and allow the generator frame ground to be connected to a ground rod separately.

A better (although much more expensive) solution is to use an isolating transformer of the full generator power rating to create a completely separate 120-0-120V supply from the 0-240, in which case the neutral and ground of both primary and secondary can be independently bonded and there is no risk of the transformer being overloaded by a heavy load imbalance or 'letting go' of the neutral.

 

[Megawatt]

@Megawatt this is not a US-spec generator, it's European spec, so it has only two wires - a neutral and a 240V hot - plus a frame ground. (We don't have 4-wire circuits or twistlocks!) The problem is that the OP has US-spec electrical wiring with a mixture of 120 and 240V appliances and panels with two 120V hots and a neutral. To run the 120V appliances and lights from the generator requires a transformer, or if the generator output winding has a hidden centre-tap, modifying its outlet box to bring that out as the neutral instead and make the existing 240V neutral into a second 120V hot.

As yet we haven't seen inside the generator we don't know whether the tap in the winding exists (it does in some units so they can be sold globally by fitting the correct outlet box). In any case it sounds like the cable to the generator location does not have the 3rd wire to allow for that modification and is difficult to replace. So the best solution so far is to send the generator's 2-wire 240V to the house with neither wire grounded (I.e. make the neutral a hot by not bonding it) and there use a transformer to create a neutral point halfway between the two, and bond that instead. A 3-pole GFCI on the transformer with the bonding connection to the transformer side of the GFCI will provide shock protection both at the generator and in the house and allow the generator frame ground to be connected to a ground rod separately.

Alternatively an isolating transformer can be used to create a completely separate 120-0-120V supply from the 0-240 the neutral, in which case the neutral and ground of both primary and secondary can be independently bonded and there is no risk of the transformer being overloaded by a heavy load imbalance. This is technically better but will be much larger and more expensive.

@Lucien my mistake I guess I thought he was from the US and I agree with you now that I know what we’re talking about. ?