Earthing Systems

Bruthal

.

Bruthal

Bruthal wrote: »

Most earth rod cables are bigger than they would need to be. A house earth electrode will never really need to be 10 square. The only good thing about it is its mechanically strong.

Right as usual Bruthal.
Earth loop impedance test passed.



RCBO working fine (trips as expected).

Photo at the back of the distro, 40m from the earth spike.

There is however a 60kVA on another end of site on a trailer with no rod to be seen.

2011

Sir Liamalot wrote: »

There is however a 60kVA on another end of site on a trailer with no rod to be seen.

Why would you need an earth rod for this generator?
I am assuming that this is an isolated electrical system (not connected to the grid) and it is an LV generator.

That's what I've been thinking 2011.
I thought it was tradition.

2011

Sir Liamalot wrote: »

That's what I've been thinking 2011.
I thought it was tradition.

"Tradition"

I was wondering what you were going to say as there is always so many opinions on earthing

Recently I was using a large generator connected to a load bank to pass 1MW through our system as part of a commissioning process. The photo below is of the controller for the load bank.

This discussion about earthing kicked off quite a debate when we were provided with a domestic earth rod for the generator earth pit. As everything was comprehensively bonded together I could not see what installing the earth rod was meant to achieve. As our system was LV the frame of the generator could be the earth. MV & HV systems are different due to capacitive coupling an earth pit is essential.

2011

2011 wrote: »

...there is always so many opinions on earthing

Double fail or higher to disaster is my preference.

2011

Sir Liamalot wrote: »

Double fail or higher to disaster is my preference.

Good philosophy.
In that case install double the amount of bonding cables.

2011

2011 wrote: »

In that case install double the amount of bonding cables.

I was thinking wiring fault + rcd/bonding fault = disaster.
Or phase to person to phase fault = disaster.

Isn't doubling up on bonding and RCDs not triple fail?



I'm in need of some plumbing advise.

I have a 2.5kVA inverter charger that I got an excellent deal on. Unfortunately it's a marine inverter not a domestic so the designed earthing of the device has some shortcomings imho.

There's a relay internally in the device for earth neutralisation to enable downstream RCDs when it's in inverter mode. In charger mode the relay is open to utilise the "shore" RCD.

I'm concerned that this relay is a potential point of failure and not failsafe.
I have been assured by the manufacturers that this relay never fails (it's nearly always no load). They're so confident that there's no fault indication for this relay failing to operate.

Below schematic is a half-arsed solution with a fault indicating neon (off when neutralised).



What I want is a uninterrupted earth to integrate into a TN-S system. I can do it with an isolation transformer...but I'd rather not (inefficient).
or I could run my secondary charger as a primary, hard wire the neutralising link on the inverter and not connect a mains input...but I'd rather not.

Any better suggestions?

The schematic doesn't have the DC connections indicated for clarity.
The reason for the 100mA RCD is there's 2 x 30mA downstream to prevent blackouts.

Bruthal

Is it the relay coil failing you're concerned about?

The coil or the switch. If the switch fails to operate so do my RCDs if it's in stand alone mode and there's no way of knowing it's failed either.

Boats don't like shore earths hence it's optional.
Easiest thing is probably bridge the earth relay 30 and 87 terminals which will probably void the warranty. I'll email the manufacturers again and ask.

If it's according to my schematic it shouldn't effect it when it's a throughput.

Can't do it externally because the inverter output is the same as the mains throughput depending on the energy source.

Bruthal

Sir Liamalot wrote: »

The coil or the switch. If the switch fails to operate so do my RCDs if it's in stand alone mode and there's no way of knowing it's failed either.

Boats don't like shore earths hence it's optional.
Easiest thing is probably bridge the earth relay 30 and 87 terminals which will probably void the warranty. I'll email the manufacturers again and ask.

If it's according to my schematic it shouldn't effect it when it's a throughput.

Can't do it externally because the inverter output is the same as the mains throughput depending on the energy source.

If RCDs are failing to operate due to no fault path, then there is no shock risk really, although it's probably better with the RCD and link setup.

A double pole relay with one doing neutral link and other the live output of invertor would show the relay is at least closed/energised.

2011

2011 wrote: »

"Why is this neutralized via a relay?
I have never seen a neutralizing point being switched.
An electrical installation should only be neutralized at one point IMHO."

I'm not a fan of it either.
It will only be neutralised at one point..at a time. Like a generator changeover.

Common practice in marine installations, galvanic corrosion boats don't want shore earths unless they're mandatory and even then mostly just ground to the case of the Isolation transformer.

Shore power; mains charging and load supply = Shore RCD.
Stand alone = inverter rcd.
Common output.

Bruthal wrote: »

A double pole relay with one doing neutral link and other the live output of invertor would show the relay is at least closed/energised.

How? What's switching the DPST relay?
The output of the inverter is sometimes the inverter sine wave and sometimes the grid throughput. Round peg, square wossits...

Bruthal

I

Sir Liamalot wrote: »

.

How? What's switching the DPST relay?
The output of the inverter is sometimes the inverter sine wave and sometimes the grid throughput. Round peg, square wossits...

Is it not the invertor output that energises the relay to close the link? So a second contact added between the neutralising relay coil supply point and the output changeover contact would need to close to supply the load via the invertor output. And once that contact is closed and supplying the load, the neutralising one must also be closed.

I'd say overall though, that the non loaded contact on the relay as it is, is more reliable than an RCD in a DB as an example. They have test buttons to periodically test them. Maybe you can add a small load between L and E with a button to test tripping of an RCD on the setup. That will show there is an earth fault path at least, unlike an RCD test button which simply bypasses one of its coils with a small internal load.

aido79

Sorry to hijack the thread but didnt want to start a new thread on earthing. I am based in Australia and was having a conversation with a couple of Australian electricians about earthing. In Australia the earthing system used is the MEN rather than the PEN that is used in Ireland. In my opinion I think the system in Ireland is safer but am looking for other people's opinions.
When I saw it first I thought it was very strange to see the earth and neutral bars linked in distribution boards but that's just the way it is here. If this link is broken or removed the consequences could be deadly.
Just wondering what other people's thoughts on this are? Which is safer?

Bruthal

The neutral from the 400v side of transformers is earthed in multiple locations here, including at each neutralised house etc.

Bruthal

Bruthal wrote: »

Is it not the inverter output that energises the relay to close the link?

That's my theory. To be confirmed. I've emailed manufacturers for a schematic and whether soldering the blighter closed is going to upset the rest of the gubbins.

Bruthal wrote: »

So a second contact added between the neutralising relay coil supply point and the output changeover contact would need to close to supply the load via the invertor output. And once that contact is closed and supplying the load, the neutralising one must also be closed.

Someat like this?






No es bueno!

The relay contacts are a mite bothersome to get to.
Attachment not found.

That's back to back with another double sided board linked with pin headers.

What I want is a permanent neutralised uninterrupted earth at the inverter the supply changeover can be the switch between neutralising points.

I accept it's rare but a working relay coil does not ensure a working relay switch.

Bruthal wrote: »

I'd say overall though, that the non loaded contact on the relay as it is, is more reliable than an RCD in a DB as an example. They have test buttons to periodically test them. Maybe you can add a small load between L and E with a button to test tripping of an RCD on the setup. That will show there is an earth fault path at least, unlike an RCD test button which simply bypasses one of its coils with a small internal load.

Yurp unloaded bog standard mechanical relays are rated for 500,000 operations iirc.

{Snip}

aido79 wrote: »

Sorry to hijack the thread

No worries. Carry on.