Inverter earth

Joined
2 Oct 2016
Messages
846
Reaction score
116
Location
Essex
Country
United Kingdom
Hi folks

It appears, from good advice received from the 'plumbing & central heating' sub forum that I need to earth my inverter in order for my combi boiler to achieve ignition, as following installation of the inverter the boiler pump runs fine but I get a E33 fault code flagged up after a few seconds.
The inverter is a CE marked pure sine wave unit with a continuous rated output of 600W & has an earth pin on the casing, but I need advice as to proceed further.

Thank you for reading.
 
Sponsored Links
The big question is can you fit the earth on the inverter, as I said having anything after the inverter that links earth and neutral could be dangerous as it could be used with another supply, the real fault is the boiler should not use the earth, that is rather naughty, but opening the inverter to connect an earth may not be easy, so looking at how to do it without danger or chance of damage, so would think first step is picture of inverter.

I would be looking at using a patch lead with earth and neutral linked some way connected to the inverter so can't be used anywhere else.

Back in the late 60's I had problems with portable traffic lights where the Mullard radar used the earth, the result was all the Mullard heads were returned and modified, today we have a problem with RCD's, DC can stop a RCD working, and I know some Bosch boilers state it must use a type A RCD not at type AC because of the DC which can stop a type AC from working. So this raises the question of the supply to a Baxi duo-tech combi 28 lpg boiler and if there should be a special RCD even without considering the inverter.

The instructions say
1. Prior to commissioning the boiler preliminary electrical system checks should be carried out.
2. These should be performed using a suitable meter, and include checks for Earth Continuity, Resistance to Earth, Short Circuit and Polarity.
The picture showing how to connect does not actually show earth being connected, and there is no reference to RCD protection, it says 2017 so one would have expected some reference to RCD protection.

A phone number is given, if it were mine I would phone and ask about connection to an IT system and what they recommend. But wait and see what others say. And tell us a little more about the inverter.
 
Thank you for your reply, unfortunately even though I've connected up the equipment safely I'm not an electrician so much of your explanation goes over my head.
As for the inverter, it is a CE marked pure sine wave unit (Mercury IPS600-12) with a continuous rated output of 600W. There is am earth connection on the outside of the casing, but the manuf. instructions only suggest that being connected to the chassis of a vehicle, presumably envisaging the unit being employed in either a motorhome or caravan rather than a domestic appliance.
 
The manual for that is as helpful as every other manual I've looked at - yes that's sarcasm, it's as helpful as a chocolate fireguard.
My guess is that internally the output is floating and just connected to the L&N contacts of the socket. The earth contact will be connected to the case and the earth stud. You'd need to open up the unit and see to be sure.
As already mentioned, it may be as simple as linking N&E inside the unit to emulate a TN supply - though personally I think it's a bit naughty for a boiler to need this as there's absolutely no technical reason to.
If you slacken the 4 screws holding the end plate on (the one with the socket), it should come away far enough to get a picture of the internal connections. If you post that, better advice might be available.

But here's a thought ...
Would linking N&E inside the inverter create a risk in itself ? As it stands, assuming it's "decently designed and built", it's an IT supply and no fault in the connected equipment can cause the inverter case to become live. Link N&E inside the inverter, and that changes - a L-true-earth fault would then cause the inverter case to become live if it wasn't earthed (say if someone borrowed it for another task).
Perhaps one option might be to fit something like a shrouded 4mm socket connected to the N - that is "reasonably" safe in that you can't touch anything without poking wires in, but would allow the use of a lead to connect N-E.
 
Sponsored Links
The manual for that is as helpful as every other manual I've looked at - yes that's sarcasm, it's as helpful as a chocolate fireguard.
My guess is that internally the output is floating and just connected to the L&N contacts of the socket. The earth contact will be connected to the case and the earth stud. You'd need to open up the unit and see to be sure.
As already mentioned, it may be as simple as linking N&E inside the unit to emulate a TN supply - though personally I think it's a bit naughty for a boiler to need this as there's absolutely no technical reason to.
If you slacken the 4 screws holding the end plate on (the one with the socket), it should come away far enough to get a picture of the internal connections. If you post that, better advice might be available.

But here's a thought ...
Would linking N&E inside the inverter create a risk in itself ? As it stands, assuming it's "decently designed and built", it's an IT supply and no fault in the connected equipment can cause the inverter case to become live. Link N&E inside the inverter, and that changes - a L-true-earth fault would then cause the inverter case to become live if it wasn't earthed (say if someone borrowed it for another task).
Perhaps one option might be to fit something like a shrouded 4mm socket connected to the N - that is "reasonably" safe in that you can't touch anything without poking wires in, but would allow the use of a lead to connect N-E.

It depends on the topology of the inverter and the isolation between the input and output.
 
From the installation manual there's an illustration which clearly shows the flame sensor as a single terminal ionisation type detector. This will sense a very low ionisation current (typically microamperes) between the boiler structure (earth) and the sensor excitation circuit when the flame is sensed.

The manual also emphasises that the installer ensures the correct "polarity" of the mains Live and Neutral and the requirement for an earth. In other words the circuit is live-neutral sensitive.

I caution against providing a hard wired link between the inverter earth and neutral on the inverter. This unlikely to permit correct operation of the boiler flame sensing circuit and, in the (unlikely) event of a DNO neutral fault could result in the inverter casing becoming live.

If the inverter has an "earth" terminal then you might try connecting this to the boiler earth terminal and firing up the boiler using the inverter as per your previous tests.
 
None of that explains why the boiler needs an earth or is polarity sensitive. Feed it a fully floating supply and no earth and tbe flame circuit can still function in the same way.

And if the inverter has a floating output - earthing it's case will have no effect.

I think we can assume that the reason for using an inverter is the absence of a mains supply. But regardkess of that, loss of supplier CNE will have no meaningful effect within the equipotential bonded zone - the inverter case could become live but there won't be a potential between it and other "earthed" items. We're talking about having the case of the inverter "bonded" to that of thd boiler, and then linking the inverter neutral to it to simulate a TN supply.
 
Once again I appreciate all the replies. It's so frustrating having laid out £300 on an inverter, battery plus incidentals only to find the the set-up won't work due to what is probably just a minor detail.
Sitting on the bog last night (I'm sure others get bright ideas in this situation) I gave this problem some thought & came up with a possible solution. There is a notice on the boiler stating 'this appliance must be earthed' & when powered by the grid it obviously is just that. However when powered by an inverter, a stand-alone device only connected to a battery, the earth wire in the 3 core cable doesn't actually 'do' anything because, unlike the grid supply, it's not grounded. I'm going to ease the front of the inverter out & see where that earth pin on the casing is connected inside ... should it be to the earth connection on the socket could simply running a suitable gauge wire from the pin to (a) the frame of the boiler or (b) the earth connection on the terminal block solve the problem?
I'll post again once I've checked the inverter wiring layout as described.
Again thanks for the continued support folks ;)

Update > Having looked inside the inverter I can confirm that the earth post on the casing IS simply linked to the earth pin of the 3 pin socket.
 
Last edited:
The problem is that the output of the inverter is PROBABLY floating. I stress the probably because they vary in design, but that's a logical way to build it. Therefore, unlike the mains where (on a TN system as almost all of us have) neutral is to a greater or lesser extent connected to earth - it's not in your inverter. IMO the boiler design is rubbish - in fact I'd go as far as saying it's incompetent as there is really zero need for that limitation.
Provided the inverter output is in fact floating, then it would be possible to connect N&E together and it would run the boiler. However, I would caution against that for the simple reason that it creates a significant potential hazard of making the inverter case live in some usage/fault combinations. For example, someone borrows the inverter for another task and plugs in some item of Class I equipment - that equipment develops a L to local earth fault which will pull the L of the inverter output to local earth. The inverter is still happily putting out 240V and so now it's L is earthed, and it's N is flapping around at 240V. Because you've connected N&E inside it, the metal case is also flapping around at 240V waiting for someone to touch it at the same time as something locally earthed - perhaps the case of the Class I equipment. Said someone is now completing a 240V circuit :eek:
The issue is that while you may understand that borrowing it and not securely earthing the earth terminal is dangerous - you can't guarantee that everyone else will understand.
That's why I suggested adding (for example) a shrouded 4mm socket to the neutral connection of the inverter - and using a separate lead to the boiler/plumbing earth. With no plug in, the inverter is floating again - and the socket is fairly finger proof. So if someone borrows the inverter, they may wonder what the extra connection is for, but there won't be the hazard that a fixed N-E link would create.

Let me tell you about something that these days when I know was was going on makes me cringe. Maaaany years ago when I were a young lad, I used to work on a farm. We had this great long extension lead that would get from one of the few sockets (I only remember there being three altogether) to pretty well anywhere we needed it. The fact that it was rubber, and many cracks had been covered in tape is not relevant - other than showing it had been around for a while. The socket was one of the old (waay back when the brand was synonymous with quality) MK surface sockets screwed on a piece of wood - and sticking out of the top was a coil of single core cable with a crocodile clip on the end. The great long length of cable was, it turned out, two core.
At this point, I think a few people reading this will immediately recognise what this arrangement was supposed to be. Just think about this - it's a two core cable with a 13A plug on one end, and a 13A socket on the other end, not normally considered a safe arrangement.
But none of us had any idea - sometimes the croc clip wasn't used at all; sometimes it was clipped the machine the lead was powering (such as a metal framed conveyor for hay bales); occasionally it was clipped to a water pipe - all screwed galv steel pipe back then; eventually it was declared a bloomin' nuisance and chopped off.
I'm sure that when the lead was made, instructions would have been given for how to use it safely. But over time, and with different people about, that information was lost. And as a result we were often using completely unearthed equipment (these were the days when a Black and Decker drill had a metal body and a 3 core flex), that was not always well maintained, and not always kept and used in dry conditions. Hopefully the danger in that is obvious.

I think many will have recognised that this arrangement was completely safe if used properly - and indeed would have been to avoid a significant hazard from using a regular 3 core extension lead. The croc clip was intended to be connected to a good local earth - such as the galv steel water pipes - and thus provide an earth that's local to the user. Using a 3 core extension lead means a) transferring an earth from one part of the premises to another at a risk of there being dangerous differences in potential during fault (and even some non-fault) conditions, and b) having a significant resistance to earth that could cause dangerous differences in potential during fault conditions (e.g. a fault in a hand tool). In extreme, b) could cause the earthed body of the drill to be at half the mains voltage until a fuse blew - which could conceivably be a while with a very long extension cable. And on a farm, you have livestock that are far more susceptible to voltage differences than we are (normally).

And that is the sort of thing that results in rules prohibiting certain things if used by unskilled people. If someone is trained in understanding the risks and how to mitigate them, then things can be allowed that you cannot trust "some random person" to use/manage safely.

But back then I knew nowt about this sort of stuff, we didn't have RCDs and stuff like that, fuses were mostly rewirable with the replacement wire based on one or both of a) what's to hand, and b) what is thick enough not to blow again. Except for the welder, one of those Pickhill oil filled jobs - that had a piece of bolt in the plug. But somehow we survived - although I recall the wiring setting on fire one day when we were pressure washing a building out, water got in the cracked rubber and it started spitting and smoking, but it was OK after drying it out and putting some tape round it :whistle: I do sometimes wonder how much of that survival was down to wearing rubber wellies most of the time :sneaky:
 
.... Therefore, unlike the mains where (on a TN system as almost all of us have) neutral is to a greater or lesser extent connected to earth ....
I'm not sure why you 'single out' TN systems - the neutral is "to a greater or lesser extent connected to earth" with any supply (even if the actual installation is TT).

Kind Regards, John
 
True, I had IT in mind.

Thinking a bit more, the OP can quickly check if the output is floating. Needs three lampholders and three (preferably) identical incandescent bulbs. Connect the lampholders to a plug so that each is across a different pair of terminals - i.e. one is L-N, one is L-E, one is N-E. When switched on, the bulb across L-N should be full brightness, while the other two should be considerably dimmer (in the order of 1/4 brightness) and equal.
Unplugging either the L-E or the N-E bulb should make the other go out, while the L-N bulb stays lit.
Unplugging the L-N bulb shouldn't affect the other two.
Use of a voltmeter would confirm things - but the bulbs are easy to interpret.
 
Once again my appreciation for all your replies guys. However because I'm basic bloke I'm having trouble understanding some of the technical advice although I fully accept that it's given in good faith.

Jackrae's advice in post #6 seems, on the face of it, to coincide with my own theory as set out in my own post #8.
Would anyone advise against at least giving this idea a try? In which case do I connect to the boiler's earth or simply to the casing/pipework?
Thanks again folks.
 
...Jackrae's advice in post #6 seems, on the face of it, to coincide with my own theory as set out in my own post #8. Would anyone advise against at least giving this idea a try?
Do you mean ..
.... If the inverter has an "earth" terminal then you might try connecting this to the boiler earth terminal and firing up the boiler using the inverter as per your previous tests.
?? If so, I can't see how that could do any harm.
In which case do I connect to the boiler's earth or simply to the casing/pipework?
If I have understood correctly, you've told us that the two are connected together (electrically), in which case it wouldn't matter which you used.

Kind Regards, john
 
A bit of wire between the inverter earth stud and any convenient bit of "earthed" metal on the boiler or plumbing would tell whether it makes a difference or not. My expectation is that it will make zero difference.
 
Thank a lot John / Simon.
I'll give 'it' a whirl over the weekend & report back :confused:
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top