Inverter earth

[SimonH2]

BTW, just out of interest, why the inverter ? Off grid ? Emergency backup ? Something else ?

 

[JohnW2]

BTW, just out of interest, why the inverter ? Off grid ? Emergency backup ? Something else ?

See here , where he says ...

.... I have installed an inverter with the intention of providing standby power to our CH boiler ...

Kind Regards, John

 

[Gasguru]

SimonH2 have you considered that having a boiler lockout in the absence of an earth connection might be seen as an added safety feature...it may actually be required by the test house. When you consider the exposed metalwork and propensity for boilers to leak ensuring earth integrity is of heightened importance.

 

[JohnW2]

SimonH2 have you considered that having a boiler lockout in the absence of an earth connection might be seen as an added safety feature...it may actually be required by the test house. When you consider the exposed metalwork and propensity for boilers to leak ensuring earth integrity is of heightened importance.

One could understand that view, but it would be much less easy to understand why it should apply to boilers if it doesn't (which, as far as I am aware, is the case) equally apply any other domestic appliances which involve electricity, particularly those which also involve water?

Kind Regards, John

 

[SimonH2]

Most boilers don't have that design flaw and I don't recall ever hearing of a spate of problems over the many many decades we've had gas boilers. As John says, why single out boilers when there are so many other devices/appliances that mix lecky and water.
Any designer who knows what he's doing could design a boiler controller that was polarity insensitive on the mains supply. And as described, the ionisation flame detector need not rely on the boiler being earthed either for operation or safety.

That they not only don't do these things, but actually put the effort into locking out, says a lot. They know they've created a completely avoidable hazard, but rather than eliminate it by design, they have used a less reliable mitigation.

 

[norseman]

Sorry for the delay in reporting gents, I've experienced various problems unrelated to the subject in hand.

OK John (post#13)... you were right, it didn't do any harm !

Simon (post#14)... you were partially right. Connecting a 2.5mm earth wire between the inverter's earth bolt & the earth pin of a 3 pin plug inserted into a nearby 13A socket didn't make the boiler function properly, however it did allow allow the boiler to fire up for a few seconds several times before the E33 error code appear, an improvement on the previous lack of any attempt to fire & an error code shortly after switching the inverter on.
I guess the next thing to try is earthing the inverter to the boiler casing itself.
Would the use of a larger gauge earth cable (I have some 6mm) make any difference, though the heavier wire would be a bit large for the terminals.

 

[SimonH2]

Simon (post#14)... you were partially right. Connecting a 2.5mm earth wire between the inverter's earth bolt & the earth pin of a 3 pin plug inserted into a nearby 13A socket didn't make the boiler function properly, however it did allow allow the boiler to fire up for a few seconds several times before the E33 error code appear, an improvement on the previous lack of any attempt to fire & an error code shortly after switching the inverter on.
I guess the next thing to try is earthing the inverter to the boiler casing itself.
Would the use of a larger gauge earth cable (I have some 6mm) make any difference, though the heavier wire would be a bit large for the terminals.

Neither of those things will make any difference - the boiler will almost certainly be earthed via the gas pipe and the main bonding, and the current needed is minuscule so even 2.5mm² wire is massively oversized in terms of making the boiler work
What you would need to do is link the neutral of the inverter to the boiler earth - having first established that it won't let the magic smoke out of the inverter. See post 11 ... The biggest issue is how to do this safely, taking into account the possibility (even if you don't plan that) of the inverter getting used for a different task.

The problem is fairly simple. For whatever reason, the boiler manufacturer has chosen a sub-optimal design and in an attempt to make that safe has added extra circuitry to detect if N & E are not connected together - most likely, simply seeing if there's any voltage between them. In the case of an incorrectly wired boiler, there could be 240V between it's N & E.
Your inverter doesn't have N & E connected together, unlike almost all mains supplies in this country. So it's earth pin will be at some intermediate voltage in betwewen it's L & N pins - the actual voltage will depend on things like stray capacitances etc. So the boiler detects this, and shuts down even though if designed sensibly it whould make no difference whatsoever to it's safety or function.
In fact, what you almost certainly have is what's called an "IT" supply - where no part of the supply is connected to earth as so it's floating. Such supplies are much safer in some situations as no single fault can make anything live. Shaver sockets contain an isolating transformer in order to create one of these IT supplies.
So basically, you have to make your inverter less safe in order to work around some idiot's idea of how to design a boiler. The whole setup would be safer if the boiler "just worked" and you could use the inverter as-is

 

[norseman]

Thank you for that Simon, what puzzles me is why my latest attempt (post #21) resulting in the boiler making several attempts to fire before showing 'error' whereas without the earth it wouldn't even try before bringing up the error code.
In my limited experience of household electrics things either work or they don't, unlike DC components in vehicles.
How about insuring that the inverter is NEVER used in any alternative scenario, there are only the two of us (adults) occupying the property & I could also affix a warning notice to the casing. Is it then a case of disconnecting the earth pin on the casing & reconnecting the wire to the neutral terminal (sharing it with the existing neutral) on the inverter's socket? Obviously I can't connect anything to the boiler terminals as they receive the 13A supply when running off the grid, using the same connecting lead.

Your (& all others) continued help with my problem is much appreciated.

 

[Aragorn84]

i would start with a simple test. Join Neutral and earth in the plug and try it out. Clearly, dont leave the link there and dont use it like that on the actual mains supply (though if its RCD protected you'll trip the RCD anyway)

If the boiler runs happily like that, then you can see about modifying the inverter to suit.

 

[SimonH2]

Thank you for that Simon, what puzzles me is why my latest attempt (post #21) resulting in the boiler making several attempts to fire before showing 'error' whereas without the earth it wouldn't even try before bringing up the error code.
In my limited experience of household electrics things either work or they don't, unlike DC components in vehicles.

The difference is likely to be because the boiler is only applying a tiny current, and there will be stray capacitance involved.
Place two conductors in close proximity and there will be both capacitive and inductive coupling between them. So if there's AC on one of them, then there will be some AC coupled to the other. When you earthed the case of your inverter, it will have created another path for some of this coupled signal to go down - and this may have been enough to change conditions enough to change the behaviour of the boiler, but not enough to completely stop the "fault" being detected.

 

[norseman]

i would start with a simple test. Join Neutral and earth in the plug and try it out. Clearly, dont leave the link there and dont use it like that on the actual mains supply (though if its RCD protected you'll trip the RCD anyway)
If the boiler runs happily like that, then you can see about modifying the inverter to suit.

Many thanks for that advice, I will do just that shortly & report back (again )

 

[norseman]

The difference is likely to be because the boiler is only applying a tiny current, and there will be stray capacitance involved.
Place two conductors in close proximity and there will be both capacitive and inductive coupling between them. So if there's AC on one of them, then there will be some AC coupled to the other. When you earthed the case of your inverter, it will have created another path for some of this coupled signal to go down - and this may have been enough to change conditions enough to change the behaviour of the boiler, but not enough to completely stop the "fault" being detected.

Your reply is appreciated, helps me understand the problems I'm having.

 

[norseman]

OK, this is where I'm at today. Removed top casing of the inverter with a view to linking the socket earth to the neutral, as suggested, only to find that the existing earth cable goes two ways ... one end to the earth connection on the casing which I've mentioned in previous post(s) but there is also a duplicate wire running to the circuit board in the inverter.
Thought I ought to run this development past you guys before proceeding further. Please see images & thanks yet again for your patience.

 

[SimonH2]

Not really surprised - it may well earth things like filters or inter-winding screen.
It doesn't really change anything except we now can't assume that the output is fully floating - it probably is, but we can't be certain. If (for example*) the design connects the earth to a centre tap of the output transformer, then linking N&E would short out the output - that "should" just blow a fuse or cause the unit to shut down, but it's not unknown for devices to just self destruct if the built in protection isn't well enough designed.
* I've seen some "interesting" ideas over the years.