Garage RCBO tripping large machinery

[JohnW2]

So i ordered a 20a hager rcbo type c after the first couple of replies but now wondering should i have lol.. Little in depth for me the replies above, Once the new rcbo get shere i'll contact a sparky and ask him to test if it will be ok, what do i need to ask him to do and why in layman's terms?

The electrician would probably welcome electrician's terms, rather than layman's ones What you need to ask him is to determine (by measurement) whether the "Earth Fault Earth Impedance" ('EFLI', also known as Zs) of the circuit is low enough to support a C20 RCBO - which, in case he doesn't know, since July 1st means that the EFLI must be no greater than 1.09 ohms.

Kind Regards, John

 

[jmac80]

Thanks very much John

 

[JohnW2]

Thanks very much John

You're welcome. Depending on how long the circuit is, I would suspect that it's probably pretty marginal as to whether the EFLI will be low enough for a C20 - so I hope you are lucky!

Kind Regards, John

 

[Risteard]

low enough to support a C20 RCBO - which, in case he doesn't know, since July 1st means that the EFLI must be no greater than 1.09 ohms.

It's considerably greater than that owing to the fact that an RCBO is an RCD.

 

[JohnW2]

low enough to support a C20 RCBO - which, in case he doesn't know, since July 1st means that the EFLI must be no greater than 1.09 ohms.

It's considerably greater than that owing to the fact that an RCBO is an RCD.

You sound like eric! As I said to him, just because there is RCD functionality (whether from a separate RCD or in an RCBO) does not mean that one is allowed to rely on that for fault protection, and nor does the RCD functionality alter the need for the device to satisfy 433.1.1(iii) as regards ('L-N') overload protection for the cable.

Kind Regards, John

 

[Risteard]

As I said to him, just because there is RCD functionality (whether from a separate RCD or in an RCBO) does not mean that one is allowed to rely on that for fault protection

But you are! There is no prohibition in the Wiring Regulations.

 

[Iggifer]

Them's the rules though! With a TN-C-S supply you have plenty of headroom to meet that figure, bit less with TN-S but not impossible. Can't see it being a problem on 10mm

 

[JohnW2]

As I said to him, just because there is RCD functionality (whether from a separate RCD or in an RCBO) does not mean that one is allowed to rely on that for fault protection

But you are! There is no prohibition in the Wiring Regulations.

Well, 415.1.2 goes some way in that direction - but, as I said, even if one is happy to rely on an RCD for fault protection, one still needs to satisfy the requirements for overload protection - and RCD functionality cannot achieve that. Are you seriously suggesting that one should ignore Zs (so long as it is less that 6,000Ω or so) when a circuit is RCD protected?

Kind Regards, John

 

[Risteard]

Are you seriously suggesting that one should ignore Zs

Of course not. But just because it exceeds the Zs for a circuit breaker does not mean that it is unsafe or fails to comply with the Wiring Regulations.

Of course you would want to verify that the Zs is appropriate to the length of cable involved.

And none of this is to suggest that I agree with relying on the RCD - I'm simply pointing out that it is permissible.

 

[JohnW2]

Are you seriously suggesting that one should ignore Zs

Of course not. But just because it exceeds the Zs for a circuit breaker does not mean that it is unsafe or fails to comply with the Wiring Regulations.

... so, in the case we are considering, where would you draw the line between 1.09Ω and 7,283Ω?

And none of this is to suggest that I agree with relying on the RCD - I'm simply pointing out that it is permissible.

It's not an approach to which I would personally want to expose DIYers.

Kind Regards, John

 

[ericmark]

There are two points.
1) With a TN-C-S system with a ELI of 0.35Ω if the line - neutral exceeds 0.92Ω then the volt drop will be exceeded so in real terms the limit is 0.92Ω.
2) Although the 0.4 seconds disconnection time is for line - earth faults. It is debatable if the 0.4 seconds actually applies to a line - neutral fault.
I realise using the type C curve for a 20A MCB it will take 30A for around 1000 seconds before it trips but it will trip before the cable melts.

 

[Risteard]

You can't have more than 7,000 Ohms as the RCD has to be able to operate with 50V or greater.

Anyhow, I stated that it has to be appropriate to the circuit length, and such a high impedance would clearly be indicative of loose connections. So work out what it should be for the length of the conductors involved, and that's what you should have.

 

[JohnW2]

1) With a TN-C-S system with a ELI of 0.35Ω if the line - neutral exceeds 0.92Ω then the volt drop will be exceeded so in real terms the limit is 0.92Ω.

True - and that could become a problem if the machine cannot function satisfactorily/safely with more than a 5% VD.

2) Although the 0.4 seconds disconnection time is for line - earth faults. It is debatable if the 0.4 seconds actually applies to a line - neutral fault.

I would say that, logically, it ought to apply, particularly in a TN-C-S installation, where neutral and CPCs are joined within the premises - in that situation, an L-N fault can raise the potential of exposed-c-ps to dangerous levels, just as would an L-E fault, can't it?

Kind Regards, John

 

[JohnW2]

You can't have more than 7,000 Ohms as the RCD has to be able to operate with 50V or greater.

Not to satisfy disconnection time requirements (which is what I thought we were talking about), it doesn't. Let's face it, when one is relying on an OPD for disconnection, the touch voltage is likely to rise to more than half the supply voltage until disconnection occurs.

Kind Regards, John

 

[westie101]

Don't forget that as far as the DNOs are concerned the ELI at the cut-out for a TNCS system can be as high as 0.8.
These figures were set many years ago by the DNOs