Inverter - TT System issue

Core Power · Oct 8, 2023

Hope this all makes as much sense as i can make of it and my "problem" can get sorted via some guru on here...

Just a quick background on myself, i am fully qualified and have been for 15+ years. I have completed the 18th edition and am a member of the approved and domestic installer schemes on NIC so not a complete newbie but i must say i am stumped and need some fresh eyes and ideas as my three brain cells are fried! :-0

Ill try and be brief. Long story short i have tidied up a static caravan that isn't going anywhere anytime soon as its remained in the same spot for 20 years or so and isnt ever moving again. (so the client says) I completed a board change (main switch and rcbo) and rectified a few c1s and c2s that i found. I carried out dead tests and all fine but here started the problem... Little did i know the supply for this caravan is going to be via solar panels, charging old forklift batteries and in turn going through a LVYUAN F-3500P Inverter. Having not much experience with this i agreed to just get the caravan up to scratch and supply a TT earthing arrangement for protection . However after driving a stake and trying to get a Ze or PFC or any type of Zs inside as well as RCBO times i cant for the life of me get 230v between line and Earth. I have since driven 2x 8feet rods (and linked them0 into the Devon countryside (its in the middle of the woods so nothing around like cables/gas/water or sewage) and cant get anywhere near the magic number. I'm starting to think i am missing something simple from the inverter side of things that maybe i don't have enough info on, Like not being grounded or inverters don't like rcbo or separate surge devices? I have never struggled to get 230v between line and a rod before. I have 230v on line and neutral coming off the 1x socket on the inverter and i get voltage of about 80v between neutral and earth. Oddly as well i get less voltage every time i turn an RCBO on which has me thinking it has to do with my supply? highest i have got is 210v which drops when main switch is turned off or rcbo is flicked on.. It really is starting to test me(Pardon the pun) i have since used about 7 stakes at various locations around the woods and starting to think it cant be as easy as trying to get a good connection on the tt side of things. I created a false tncs by a link between the earth and neutral bar in the board to see if i could get anything and managed to get some results but obviously wont and haven't left it like that. i just cant help but think im missing something and have no problem in accepting that. Hope someone can help. Thanks for reading.

pc1966 · Oct 8, 2023

Fundamentally, measuring Ze (or indeed Ra) by using the supply is making the assumption the supply originates from something with a low impedance to the true Earth in the first place. If you have a locally generated supply (common for generators, etc) it is usually TN-S and the rod is simply referencing N to Earth, it is not TT as there is no far-away rod forming the other side of the E path.

Also you need to check if the inverter has a N-E link internally on the AC output, or if it is generating a floating (IT) supply.

To measure the rod you would need to use an MFT that supports spikes or similar to get an idea of the impedance, or if you have two rods separated by a few lengths you can earth via one rod and measure via the other, the true impedance will be less than the rod-rod (assuming, again, they are separated by at least couple of rod's lengths so not bound up in each other's immediate region of influence).

Core Power · Oct 8, 2023

pc1966 said:
Fundamentally, measuring Ze (or indeed Ra) by using the supply is making the assumption the supply originates from something with a low impedance to the true Earth in the first place. If you have a locally generated supply (common for generators, etc) it is usually TN-S and the rod is simply referencing N to Earth, it is not TT as there is no far-away rod forming the other side of the E path.

Also you need to check if the inverter has a N-E link internally on the AC output, or if it is generating a floating (IT) supply.

To measure the rod you would need to use an MFT that supports spikes or similar to get an idea of the impedance, or if you have two rods separated by a few lengths you can earth via one rod and measure via the other, the true impedance will be less than the rod-rod (assuming, again, they are separated by at least couple of rod's lengths so not bound up in each other's immediate region of influence).

Thanks for the reply. It does say on the inverter it’s a “floating neutral”. Yea I was thinking of going down this route with regards to the rods but something in me is saying I can lay as many rods as I wanted or Matt’s, tapes etc but I won’t achieve the 230v but could be wrong. No other supply at all on the premises. Only power is from the solar panels and the forklift batteries into the inverter.

davesparks · Oct 9, 2023

If it has a floating neutral then the supply is IT as it stands and no amount of buried metal will change that.

If the installation is unsuitable for an IT supply (no insulation monitoring, not under the supervision of a competent person etc) then you need to consider how you can convert it to TNS using earth rods to reference the neutral to earth.

Core Power · Oct 9, 2023

davesparks said:
If it has a floating neutral then the supply is IT as it stands and no amount of buried metal will change that.

If the installation is unsuitable for an IT supply (no insulation monitoring, not under the supervision of a competent person etc) then you need to consider how you can convert it to TNS using earth rods to reference the neutral to earth.

Thanks Dave, I appreciate the reply as I knew it was something I wasn’t picking up.I’ll read up in the regs book on an IT supply as Can’t say I’ve ever come across the need for one. Looks like I’ll have to look into converting it to a TNS. Thanks again. Any more info on this would be a great help but you’ve really pointed me in the right direction. Thanks again