Island Mode N-E Bond Relay, How do inverters manage this?

[SafetyThird]

When I was installing my Luxpower inverter and batteries, my electrician came across an issue where the system doesn't behave correctly in island mode, ie when the grid is out and we're running from batteries. Rather than try and explain it myself, he's written it out in detail because we need to forward it on to Luxpower. I thought there's bound to be someone on here who's come across this issue before and might be able to either shed light on how to fix it.

Hi J

As per our conversation today, I believe there is an issue with the inverter switching between Connected Mode and Island Mode (switching between on and off-grid).

When in connected mode the voltages at the incoming terminals of the Critical Supplies consumer's unit connected to the UPS output are L-N 230V, L-PE 230V and N-PE 0V. This is as expected.

When simulating mains power failure, the inverter switches into UPS mode. My understanding is that the inverter includes an internal "Island Mode Isolator" that isolates the incoming grid L & N from the outgoing UPS L & N within the inverter when running in Island Mode. This is as required by UK Regulations when any generator runs in Island Mode.

In Island Mode I obtained readings of 230V between L & N, but the voltage between L and PE and N and PE varies wildly and I measured as much as 90V N to PE at one point.

Below is a link to the IET website (Institution of Engineering and Technology, the organisation that writes the Wiring Regulations and other electrical publications). It links to excerpts from their Code of Practice for Electrical Energy Storage Systems and details the connection requirements for Connected Mode and Island Mode and specific requirements for switching between them.

Reference is made to the need for a N-E Bond Relay. This relay contact switches a link between the N and PE of the UPS output dependent on the status of the Island Mode Isolator; as the Island Mode Isolator throws open (under grid power loss), the N-E contact closes thus grounding the N of the UPS supply from the inverter and preventing a "floating neutral". When the Island Mode Isolator makes again (on restoration of the grid supply), the N-E contact opens thus removing what would be detected as a N-PE fault by any upstream RCD and preventing it from tipping.

I can find no reference to a N-E Bond Relay in the documentation for the inverter. There is no reference to an internal one nor a reference for the need for one to be installed externally, however, it is clear from the document in the link that one is required.

From the voltages I have obtained I believe there is no N-E Bond relay present within the inverter.

Due to the requirement for the relay contact to close after the isolator has opened, but open before the isolator closes I believe the relay needs to be controlled by the inverter; the inverter is in charge of when switching takes place.

The questions I would put to Lux Power are:
Is there a N-E Bond Relay integral to the inverter and if so why am I obtaining the readings detailed above?
If there is not a N-E Bond Relay integral to the inverter, how can one be installed and connected externally?
Depending on the answers to 1 and 2, why is there no reference to the need to install a N-E Bond Relay or instruction on how to connect one to the inverter within the manual.
https://electrical.theiet.org/wiring-ma ... e-systems/

Hopefully I have provided sufficient information for Lux Power to understand the nature of the issue and the query, but if they require any further information I would be happy for them to contact me.

Hopefully they will be able to provide us with a solution to this problem.

Regards

Martyn

[Swwils]

Good luck with that request lol.

They do mention this in the hybrids manual as "mode A / mode B".

There are also several hidden settings, these are formatted from my HA integration:

switch.lux_seamless_eps_switching
switch.lux_anti_island_enable
- switch.lux_dci_enable
- switch.lux_drms_enable
- switch.lux_feed_in_grid
- switch.lux_gfci_enable
- switch.lux_grid_on_power_ss
- switch.lux_neutral_detect_enable
- switch.lux_ovf_load_derate_enable
- switch.lux_r21_unknown_bit_12
- switch.lux_r21_unknown_bit_3

[nowty]

There have been many threads about this and a lot of disagreement, the last one which you were even party to (at least at the start) is here, viewtopic.php?f=14&t=1401
so I'm only going to say the minimum which should not be too controversial and this is the last I will say on the subject in this thread.

Each country has different rules and regulations regarding electrical installations and earthing so its practically impossible for inverter manufacturers to be compliant to all of them.

Generally, the output of EPS in inverters is typically a floating voltage and only used in an emergency. So, it's safe'ish to run an extension scocket and only use simple devices which are double insulated which don't use earths like phone chargers, laptops, internet routers, some simple table lamps. I say "safe'ish" because once you run an EPS extension socket its hard to prevent people plugging in what they want.

If you want to use an appliance like a kettle, toaster, fridge, microwave, etc, (with an earth), to be safe then it gets complicated and can become expensive.

Your electrician is correct that there needs to be,

• a switching mechanism to disconnect live conductors of the installation that are to be powered in island mode from the grid. The IET Code of Practice for Electrical Energy Storage Systems calls this an island mode isolator

• a switching mechanism to provide a neutral for the island mode The IET Code of Practice for Electrical Energy Storage Systems calls this an N-E bond relay

But he has not mentioned (yet) the third requirement,

• a consumer earth electrode

Its not clear from the LUX manuals exactly how their hardware deals with the first two. My SMA Sunny Island gives me detailed information of how to accomplish it but most of it (isolation and switching) is done externally to the battery inverter.

[AE-NMidlands]

I am a bit disturbed by all the discussions about earthing in different circumstances. Is there any way that too many earths can do harm? Cross-bonding gas and water pipes (if they are continuous with metal underground) can give you extra earths anyway.

Given that they (earth spikes) don't sound very expensive to install why not have as many as you could possibly need? The one bit I suspect might go wrong is if your "mains" earth is connected to the supply neutral, you are islanded and being powered by batteries and you have a fault which puts a voltage on your joint earth and hence on the mains neutral.

If this is right, surely it can be dealt with by having an isolator on the connection to the "mains" earth too, which is broken when the mains is off? Although that sounds like a bad idea in principle.

[Swwils]

I am leaning towards in the default configuration the lux unit has a anti islanding relay but NOT N-E bond relay.

Which means your island mode isolator (usually a change over switch) will also need a contactor or something to do this for you. Since I cannot find any way to get a EPS signal out of the lux box.

Other brands either have a N-E relay, a EPS signal out for a external relay, or require a contactor for EPS anyway.

I did have full internal PCB pictures somewhere but I can't find them.

[Stinsy]

I am a bit disturbed by all the discussions about earthing in different circumstances. Is there any way that too many earths can do harm? Cross-bonding gas and water pipes (if they are continuous with metal underground) can give you extra earths anyway.

Given that they (earth spikes) don't sound very expensive to install why not have as many as you could possibly need? The one bit I suspect might go wrong is if your "mains" earth is connected to the supply neutral, you are islanded and being powered by batteries and you have a fault which puts a voltage on your joint earth and hence on the mains neutral.

If this is right, surely it can be dealt with by having an isolator on the connection to the "mains" earth too, which is broken when the mains is off? Although that sounds like a bad idea in principle.

Interestingly up until the very latest version of the regs, you were expressly prohibited from installing your own earth rod in parallel with he supplier earth. Now with the very latest version (it seems*) you are required to fit your own earth rod in parallel with the supplier's earth.

* The regs have alway been poorly worded and vague. They are also filled with contradictions and use definitions for words that are different from the everyday meaning. The long awaited "Guidance Note 8" will give clarity on earthing requirements (hopefully).

However the point is to make installations "safe". As with any complex set of rules, they are "For the guidance of wise men and blind obeyance of fools". The regs are expressly not supposed to be obeyed blindly, an installer is supposed to apply judgement.

[Swwils]

Depending on how you read and interpret it some say you meant to install a customer earth rod on any electrical works.

Imagine the chaos! And now everyone awaits the guidance note.

Didn't they try to get foundation earthing going aswell? Just to realise how are you meant to practically time a sparky being on site for that.

Maybe once we have confirmed information about the hardware and software lux unit, we could start a pinned mega thread with the information and example diagrams.

I am struggling to find a spark even remotely interested.

[Oldgreybeard]

There's no requirement to install an additional intermediate on the customer side of a PME installation, the only change in the regs is that it is now allowable. The reality is that it has always been "allowable", for example, about 1m below our meter kiosk there is a pot joint where our single phase 35mm concentric incomer connects to a three phase underground run of Wavecon95. That pot joint has the usual length of copper earth strap coming out of it and laid along the bottom of the trench as an intermediate. Slight nuisance from my perspective, as it meant that I had to position our TT earth electrode outside the zone of influence of that intermediate.

The IET have lost it, IMHO. There are now so many conflicting and confusing parts of the regs that they really need to do a root and branch reform and make a proper safety case for every recommendation in the regs. At the moment, the weight placed on low probability risks is far to high, and there is a need to get back to a common sense approach. Foundation earthing is a good example. Everyone was of the view that this would come back as an option, perhaps even be mandated for new builds in the last amendment, alas it got left out, probably because of lobbying from developers. We used to use foundation earthing for steel framed buildings, decades ago, but the IET decided in their wisdom to bin it, as they were focussed on the magnificence of PME as the solution to all issues. The fact that PME both carries its own significant risks, and that it doesn't work in many rural areas, seems to have been given a stiff ignoring.

[SafetyThird]

I'll come back in the morning and read through this in detail when I'm less tired. Thank you all for the comments so far. I should add, my electrician installed an earth rod for the battery setup.

[vantech2022]

Just to jump on this, the Alpha ESS battery systems operate in this way (at least the Smile B3).

When in backup mode, i can measure 120V between L&E and N&E and 230V between L&N. I contacted them, they told me it was because of capacitors between the various connections when in backup mode.

I asked about adding a N-E bond, and it was a firm “do not do this”. I raised this with their UK distributor, who told me that the backup ideally needs to supply either a double socket or a separate consumer unit (as is my case) with an earth rod outside (which I also have to supplement the PME).

If you compare this to Tesla and others, their backups are not UPS’ as such, in that they fail over for a few seconds and this N-E bond is created. I would add that personally my view, as this is for backup purposes only, there is little risk (I am feeding a socket and my lighting circuits). I am sure this would open a debate elsewhere, however regulations are constantly changing, look at EV chargers! I am keeping an eye on this subject.

Alpha didn’t really seem to understand, and I actually purchased the inverter/battery having checked with them beforehand, and I was given the wrong information (they told me it created a TN-S output from memory, when in actual fact (after testing / install) it appears to be more aligned with an IT type earthing system).