High(ish Zs and extending ring final question.

[Marcus Vaughan]

Hi all. Love the forum by the way.

I do have a question though...

I have just visited a house (on a Sunday!). Customer is asking for additional sockets onto 32amp ring final. Circuit is 30mA RCD protected. Wiring is 2.5mm/1.5mm tinned copper.

Thought I’d better do a live Zs test.

Zs = 1.29, down to 1.22 back of socket. Checked a nearby socket - same results.

So it doesn’t comply with Max Zs values for ADS and I am not one of those who likes to use the ‘RCD get out of jail card’.

So by extending it I will be knowingly making the circuit less compliant.

Impractical to split circuits (no room in the CU). Impractical to rewire in larger cable.

Feels like a no-go.

I was wondering if, with those values, it would make enough difference if I conducted the dead test for more accurate results?

Clutching at straws.

Flooring going in Wednesday apparently....

Irony is that I know they will just use extension leads across the floor instead.

I wonder what you would do?

 

[Octopus]

You should properly dead test the circuit before you can make any decisions - starting with the Ze and then the end to end readings of the circuit..

If those are the true readings, maybe consider dropping the breaker to 20A .....

 

[SparkyChick]

What he said ^^^

 

[Andy78]

Do your dead tests and calculate the Zs and compare with live readings. If the calculated is much lower you may have a high resistance somewhere, for eg I have heard some RCDs can do this.

 

[Marcus Vaughan]

Thanks for quick replies. I had something similar before. I managed to change the MCB to 25amp. But then worrried that I had compromised on 433.1.204 - that is to say that ring finals MUST be protected by 30 amp or 32amp.

Reading it through again - I see the words ‘may be supplied...’. So technically we are not breaking the reg.

Are we?

No issues with loading.

 

[Marcus Vaughan]

Do your dead tests and calculate the Zs and compare with live readings. If the calculated is much lower you may have a high resistance somewhere, for eg I have heard some RCDs can do this.

Hi Andy, I was hoping you would chip in. Thank you.

That makes sense to me. The circuit doesn’t seem particularly long. The RCD is suspicious - tripping on live tests. Though RCD tested OK - including Ramp Test.

I think I would prefer to at least know that I can change the MCB if I need to - without falling foul of regs.

 

[Octopus]

You need to do the Ze and dead tests .....

 

[Mikeluker]

Hi all. Love the forum by the way.

I do have a question though...

I have just visited a house (on a Sunday!). Customer is asking for additional sockets onto 32amp ring final. Circuit is 30mA RCD protected. Wiring is 2.5mm/1.5mm tinned copper.

Thought I’d better do a live Zs test.

Zs = 1.29, down to 1.22 back of socket. Checked a nearby socket - same results.

So it doesn’t comply with Max Zs values for ADS and I am not one of those who likes to use the ‘RCD get out of jail card’.

So by extending it I will be knowingly making the circuit less compliant.

Impractical to split circuits (no room in the CU). Impractical to rewire in larger cable.

Feels like a no-go.

I was wondering if, with those values, it would make enough difference if I conducted the dead test for more accurate results?

Clutching at straws.

Flooring going in Wednesday apparently....

Irony is that I know they will just use extension leads across the floor instead.

I wonder what you would do?

With the earth loop reading, this would be fine with an RCD fitted, take a LN loop test to see if it meets short circuit protection for the device being used, if not just down rate the protection device to suit.

 

[Octopus]

With the earth loop reading, this would be fine with an RCD fitted, take a LN loop test to see if it meets short circuit protection for the device being used, if not just down rate the protection device to suit.

Not sure if I agree with this ..... I would say further investigation IS required .... relying on RCD's isn't a "get out" IMHO

 

[SparkyChick]

Not sure if I agree with this ..... I would say further investigation IS required .... relying on RCD's isn't a "get out" IMHO

Totally agree with @Murdoch . Simply saying 'drop the MCB rating' or 'rely on the RCD for fault protection' isn't good enough.

Dead tests should be carried out (as already stated) and used to determine where the cause of the high Zs is. Could be the earth for the whole installation is bad requiring DNO intervention, or a bad connection somewhere on the RFC.

And no, you're not breaking the regulations by using a lower rated MCB for an RFC. As the designer you can elect to do that.

 

[Marcus Vaughan]

And no, you're not breaking the regulations by using a lower rated MCB for an RFC. As the designer you can elect to do that.

Thank you for that. It is an old GEC vynckier so I might (will) struggle....

 

[Octopus]

Thank you for that. It is an old GEC vynckier so I might (will) struggle....

Ebay or ask on here ...

GEC VYNCKIER SERIES S MCB 2A 6A 10A 16A 20A 32A 40A S2 S6 S10 S16 S20 S32 S40 | eBay - https://www.ebay.co.uk/itm/GEC-VYNCKIER-SERIES-S-MCB-2A-6A-10A-16A-20A-32A-40A-S2-S6-S10-S16-S20-S32-S40/281500228641?hash=item418ab89c21:m:mFPaHZMnHgkEy8T0kEklyGQ

 

[Marcus Vaughan]

Yep. Just seen one on ebay. 25 amp.
Not a huge fan of 2nd hand breakers but this is probably the lesser of all evils.

I think I’ll try and persuade the customer that I should dead test and if still necessary search for anything that might be causing the readings. I should also warn him that I might need to change the MCB.

Sounds like a plan.

A bit short of time though....

Pressure is on!

Ebay or ask on here ...

GEC VYNCKIER SERIES S MCB 2A 6A 10A 16A 20A 32A 40A S2 S6 S10 S16 S20 S32 S40 | eBay - https://www.ebay.co.uk/itm/GEC-VYNCKIER-SERIES-S-MCB-2A-6A-10A-16A-20A-32A-40A-S2-S6-S10-S16-S20-S32-S40/281500228641?hash=item418ab89c21:m:mFPaHZMnHgkEy8T0kEklyGQ

 

[Madocks]

What earthing system?

 

[Marcus Vaughan]

What earthing system?

Didn’t look to be honest. Just a short visit on a Sunday’s. Almost certainly TNS though, all other properties around here are.