RCD's What do you use What do you think

[b.james]

On my inverter 240 volt output I wire it through an RCD (Residual Current Device) then on to a power point . The thought simply being that all of my grid power these days goes through them in case someone sticks a fork in an outlet.

I didn't see anything said by the inverter maker about this and I was not going to short it out to test it .
So what do you use and what do you think ?

Wiki definition in depth here.

RCDs operate by measuring the current balance between two conductors using a differential current transformer. This measures the difference between current flowing through the live conductor and that returning through the neutral conductor. If these do not sum to zero, there is a leakage of current to somewhere else (to earth/ground or to another circuit), and the device will open its contacts. Operation does not require a fault current to return through the earth wire in the installation; the trip will operate just as well if the return path is through plumbing, contact with the ground or any other current path. Automatic disconnection and a measure of shock protection is therefore still provided even if the earth wiring of the installation is damaged or incomplete.

This means that if a rat chews through a wire insulation and shorts it or some water gets in it will operate.

 

[offgriddle]

I wish I could help but you left me with more questions than answers! First I've heard of an RCD, kinda wondering if I should have one! Also, my next inverter upgrade will be to a 220 output to run my well pump, what kind of inventor do you have? And are your outlets at 240 volts and if so is having such high voltage in the outlets normal in the land down under?

 

[gnubie]

RCD, GFCI, same thing.

They all require that there is a ground return for the current otherwise there can be no current flow outside of the active(line/live if you prefer) and neutral wires to trip the device. That also means it's safe to hang onto the active wire if you aren't also connected to neutral (do not try this!). It is very hard to ensure that there is never a ground reference in a system so the rule is make it grounded and then make provisions for something going wrong.

Many, far too many, inverters are not configured in a way that this would allow an RCD to function. The chasis ground point on the inverter, which may also be connected to the ground pin on a GPO if the inverter has sockets, isn't necessarily connected in any way to neutral so no current from the active wire can flow via that ground pin. Neutral / earth bonding regs are different for each country so I won't discuss it in this forum. However there are inverter designs out there where if the neutral is tied to ground the inverter will self destruct.

 

[offgriddle]

RCD, GFCI, same thing.

They all require that there is a ground return for the current otherwise there can be no current flow outside of the active(line/live if you prefer) and neutral wires to trip the device. That also means it's safe to hang onto the active wire if you aren't also connected to neutral (do not try this!). It is very hard to ensure that there is never a ground reference in a system so the rule is make it grounded and then make provisions for something going wrong.

Many, far too many, inverters are not configured in a way that this would allow an RCD to function. The chasis ground point on the inverter, which may also be connected to the ground pin on a GPO if the inverter has sockets, isn't necessarily connected in any way to neutral so no current from the active wire can flow via that ground pin. Neutral / earth bonding regs are different for each country so I won't discuss it in this forum. However there are inverter designs out there where if the neutral is tied to ground the inverter will self destruct.

Aah GFI, yep, gotcha I suspected possibly thats what he was referring to as an RCD thanks gnubie! Now ... GPO?

 

[gnubie]

general purpose outlet, aka power point, wall socket.

 

[b.james]

RCD, GFCI, same thing.
They all require that there is a ground return for the current otherwise there can be no current flow outside of the active(line/live if you prefer) and neutral wires to trip the device.

I think not ..See the OP I put a quote there . It does not connect to earth. It measures current out to current in . Your post I think has just clouded the issue and I wonder if you might revise it please. I'd like to discuss ,and stick to ,the facts .

 

[gnubie]

RCD and GFCI both check for imbalance between current flowing in active and neutral. For there to be another path the ground / earth must be tied to neutral at some point UPSTREAM of the RCD of course otherwise it won't sense for the reason stated already.

The typical configuration is a hair trigger mechanism with an electromagnet that is operated from the current produced by a basic transformer (a ferrite ring with active and neutral wrapped in such a way that their sum current is nil, with the electromagnet it's source also a winding on that transformer. There are devices with electronics in them to improve the reaction time of the device, but the basic principle is the same.

Of course, if you care to present any evidence that convinces me to the contrary I'll proudly stand up and admit error.

 

[Rider]

RCD and GFCI both check for imbalance between current flowing in active and neutral.
...

Of course, if you care to present any evidence that convinces me to the contrary I'll proudly stand up and admit error.

This is how I understand it too. It is not checking for a fault in the ground as the name kind of implies, but rather a difference in the current between hot and neutral and a potential incorrect path from hot to ground. A fault TO ground, not a fault IN the ground.

 

[gnubie]

This means that if a rat chews through a wire insulation and shorts it or some water gets in it will operate.

No it doesn't. You can short the active and neutral all you like. The RCD will not trip. The rat needs to provide a return path that is not the neutral wire. It could bridge active and earth wire, and that would trip. It could short active and earth and that would trip, assuming the fuse didn't go away first. There has to be an imbalance in the current flowing in the active and neutral wires and for that to happen there has to be a return path that is not the neutral wire.

If the grounding system is not tied to neutral upstream of the RCD there is no return path that will cause the device to trip.

The only thing apparent here is that you did not actually understand my post regarding the function of a RCD / GFCI in post #3. This is why I won't discuss the specifics of earthing on this forum.

 

[b.james]

No I'm not understanding because its not what I have learnt to date . That's why I am asking the questions . If a short happens between active and neutral you say it will not trip it but the fuse will trip . Right?

OK if that is so then getting back to an RCD fitted to an inverter on my solar system needs what? Are you saying I should earth my inverter to the house earthing which is used by the grid tie lets say on the same earthing rod ? My panels are already earthed to that point so there should be no damage right .

I'm looking to be sure that the RCD provides protection . That is the main point . How do I do that?

 

[gnubie]

No I'm not understanding because its not what I have learnt to date . That's why I am asking the questions . If a short happens between active and neutral you say it will not trip it but the fuse will trip . Right?

No. Stop misrepresenting what I said. What I said in this matter is very clear and concise.

OK if that is so then getting back to an RCD fitted to an inverter on my solar system needs what? Are you saying I should earth my inverter to the house earthing which is used by the grid tie lets say on the same earthing rod ? My panels are already earthed to that point so there should be no damage right .

I will not advise on matters specifics of earthing systems on this forum as I have already stated. If you have any doubts obtain the electrical standards for your state and read them.

 

[b.james]

. Neutral / earth bonding regs are different for each country so I won't discuss it in this forum. However there are inverter designs out there where if the neutral is tied to ground the inverter will self destruct.

This is the bit that riles me . It is vague and it borders on being wowserism. We are both in Australia thats clear . Lets answer it in an Australian way. Let your answer be related to Australia only if you wish and labeled clearly ,but I seek an answer .

If you are not willing to answer then why are you posting in the thread?

 

[gnubie]

On my inverter 240 volt output I wire it through an RCD (Residual Current Device) then on to a power point . The thought simply being that all of my grid power these days goes through them in case someone sticks a fork in an outlet.

I didn't see anything said by the inverter maker about this and I was not going to short it out to test it .
So what do you use and what do you think ?

Wiki definition in depth here.



This means that if a rat chews through a wire insulation and shorts it or some water gets in it will operate.

There is no mention of discussing specifics of earthing systems in your opening post. You were querying what people used and thought, presumably about the use of RCDs.

You disputed my understanding of RCDs, retro fitted a wikipedia link to your opening post which, incidentally, completely supports what I have said about RCDs and have repeatedly called into question my understanding of RCDs yet you still do not state exactly what you believe to be incorrect about my understanding of the devices.

 

[Johnny]

I didn't see anything said by the inverter maker about this and I was not going to short it out to test it .

Hi B. James, just a thought my RCD's have a test button, I don't know if this is a good enough test through an inverter though.

John

 

[Rider]

Hi B. James, just a thought my RCD's have a test button, I don't know if this is a good enough test through an inverter though.

John

Well, I came across something interesting with testing, and it left me more confused. My inverter does not appear to have a GFCI. I have an external GFCI tester. Insert it into the outlet and press the button. If everything is working properly, the GFCI (if present) will trip. With my inverter, it did not trip. Ok, so I added an external GFCI outlet and wired it in correctly. It has a test button. When I push the test button with power applied, it trips fine. If I plug in my GFCI external tester and press it's test button, it will not trip. So I'm really not sure GFCI is working correctly. And yes, if I use the external GFCI tester on a GFCI protected circuit in my house, it trips a GFCI protected circuit correctly.

 

[offgriddle]

general purpose outlet, aka power point, wall socket.

AH Hah .. thank you for the clarification once again!

 

[offgriddle]

There is no mention of discussing specifics of earthing systems in your opening post. You were querying what people used and thought, presumably about the use of RCDs.

You disputed my understanding of RCDs, retro fitted a wikipedia link to your opening post which, incidentally, completely supports what I have said about RCDs and have repeatedly called into question my understanding of RCDs yet you still do not state exactly what you believe to be incorrect about my understanding of the devices.

breathe

 

[offgriddle]

This is the bit that riles me . It is vague and it borders on being wowserism. We are both in Australia thats clear . Lets answer it in an Australian way. Let your answer be related to Australia only if you wish and labeled clearly ,but I seek an answer .

If you are not willing to answer then why are you posting in the thread?

breathe

 

[offgriddle]

~~ Ahhhh yes .. the age old, "which came first, the ground or the neutral", conundrum. Well, even electrical engineers and theorists do not fully understand how ground and neutral can be the same, yet different, how neutral and ground can be tied together at the same point sometimes, yet at other times require separation, there is simply a certain amount of accepting that, "it is what it is"!
~~ With the afore mentioned in mind, we know that a RCD or GFCI protection device operates on the principle of monitoring the imbalance of current between the circuit’s ungrounded (hot) and grounded (neutral) conductor. An interesting point about these devices is that despite their name, they will operate on a circuit with or without an equipment grounding conductor.
~~ Me, I just use fuses in DC mixup and when it comes to feeding a main breaker panel with an AC inverter, I bond the neutral and earth ground together in the main panel and separate the neutral and earth ground in any sub panels.
~~ Why separate the neutral and ground in subpanels? If you have the neutral and ground bonded at a subpanel, then you'll get neutral return current through the ground wire back to the main panel (since there are now multiple paths). Even worse if the neutral ever has a fault, everything will continue to work but you'll have all the current on the ground, which also means that you can now be electrocuted by touching the panel chassis, for example, peace.

 

[gnubie]

Ok, so I added an external GFCI outlet and wired it in correctly. It has a test button. When I push the test button with power applied, it trips fine. If I plug in my GFCI external tester and press it's test button, it will not trip. So I'm really not sure GFCI is working correctly. And yes, if I use the external GFCI tester on a GFCI protected circuit in my house, it trips a GFCI protected circuit correctly.

The test button on a RCD (GFCI) is only there to test that the device itself can react to an imbalance in the active/line/live and neutral current. It does this by, typically, hooking a resistor between active and neutral across the sense transformer so that there will be an imbalance in the current flowing the active and neutral winding in the sense transformer. That's all it does. It is not testing any other part of the wiring and earthing system. That it trips is no indicator at all of the overall safety of the entire system.

A test plug that you put in a socket to see if it will trip the RCD is again only a partial test. It only shows that the neutral and ground are bonded upstream of the RCD under test and that the RCD can detect the imbalance caused by passing current around it via the earth wire. Its important to understand that this is not a full test of the installation. If the earth system is not to spec as per local regulations the RCD can not offer the full level of protection it normally would.

*edit to add additional comment below*
The tester connects the active/line/live wire to the earth wire with a resistor. The reason the external tester does not trip your GFCI that is connected to the inverter is that the earth point (and earth pin in the inverters output socket) on the inverter is not tied to neutral in the inverter, and not externally either. No current can flow across active and earth to cause the imbalance so the GFCI will not trip. This is something I mentioned in an earlier post.