I'm confused about junction box rules

[SUNRAY]

The figures for the 1/8 rule are there - albeit assuming a ring end to end resistance of 32Ω (corrected from my previous editing error)
Scaling that to 100m of 2.5mm² or 741mΩ, lets approximate that to 800mΩ for simplicity for a factor of 32/0.8 = 40

Showing LN loop resistance at origin, 50%, 62.5%, 75%
1/8 rule cable length for spur
length of 2.5mm² spur to achieve 0.4Ω

Happy to have errors pointed out, several times during calculations I confused R1 + R2 with just R1

 

[JohnW2]

I think that is where we are all going wrong. ... Stop thinking about it and just throw it away.

Maybe, but that's not a very intellectually satisfying approach! Also, I'm not going to throw it away, since some of the stuff in its Appendices can be quite useful!

However, I think this part of the discussion has served one useful purpose, by reminding people of something that often seems to be overlooked. We see a lot of questions, answers and discussion about spurs from ring finals - where they are permitted, what they may supply, where they should originate, what size cable etc. etc., but we very rarely see and mention or discussion about the fact that the Zs at the end of the spur obviously should not exceed the 'maximum Zs' for the OPD protecting the circuit.

As I've hinted, I don't currently believe that the OSG's 'rule-of-thumb' is necessarily a satisfactory or appropriate way of ensuring that the 'maximum Zs' is not exceeded, but the discussion reminds us that one really should confirm that such is the case (by measurement and/or calculation).

 

[SUNRAY]

Maybe, but that's not a very intellectually satisfying approach! Also, I'm not going to throw it away, since some of the stuff in its Appendices can be quite useful!

However, I think this part of the discussion has served one useful purpose, by reminding people of something that often seems to be overlooked. We see a lot of questions, answers and discussion about spurs from ring finals - where they are permitted, what they may supply, where they should originate, what size cable etc. etc., but we very rarely see and mention or discussion about the fact that the Zs at the end of the spur obviously should not exceed the 'maximum Zs' for the OPD protecting the circuit.

As I've hinted, I don't currently believe that the OSG's 'rule-of-thumb' is necessarily a satisfactory or appropriate way of ensuring that the 'maximum Zs' is not exceeded, but the discussion reminds us that one really should confirm that such is the case (by measurement and/or calculation).

Maybe but if my calculations are correct it works well for the middle half of the ring

 

[JohnW2]

Happy to have errors pointed out, several times during calculations I confused R1 + R2 with just R1

Thanks. As I said, I'll look into this properly when I have a little time, probably this evening.

However, in the meantime, what does your approach say about the maximum permissible spur length for a spur originating very close to the origin of the ring? - as I said, a small amount of thought suggests that (assume ring midpoint is already at 'maximum Zs') a spur could be half of the distance to mid-point of ring, not one-eighth of that distance.

 

[EFLImpudence]

but we very rarely see and mention or discussion about the fact that the Zs at the end of the spur obviously should not exceed the 'maximum Zs' for the OPD protecting the circuit.

Why would we? It goes without saying - the same as any other circuit.

 

[JohnW2]

Why would we? It goes without saying - the same as any other circuit.

It certainly should do but, as I said, thinking back, I can't recall ever having noticed the issue being raised in the (many) discussions about spurs.

It's pretty common for people to come here asking things like "can I take an unfused spur from socket A to feed a new socket/whatever B". The answers usually major on "provided socket A is 'on the ring' and not already a spur" and "provided it feeds only one socket/FCU/whatever" but I can't say I recall that being qualified by ".... provided, of course, the Zs at the end of the new spur is not above the 'maximum Zs' permitted in relation to the circuit's OPD". Do you recall that often (if ever) being added ?

 

[pete01]

It certainly should do but, as I said, thinking back, I can't recall ever having noticed the issue being raised in the (many) discussions about spurs.

It's pretty common for people to come here asking things like "can I take an unfused spur from socket A to feed a new socket/whatever B". The answers usually major on "provided socket A is 'on the ring' and not already a spur" and "provided it feeds only one socket/FCU/whatever" but I can't say I recall that being qualified by ".... provided, of course, the Zs at the end of the new spur is not above the 'maximum Zs' permitted in relation to the circuit's OPD". Do you recall that often (if ever) being added ?

What's the point most people haven't got a tester to measure it anyway.

 

[ebee]

It certainly should do but, as I said, thinking back, I can't recall ever having noticed the issue being raised in the (many) discussions about spurs.

It's pretty common for people to come here asking things like "can I take an unfused spur from socket A to feed a new socket/whatever B". The answers usually major on "provided socket A is 'on the ring' and not already a spur" and "provided it feeds only one socket/FCU/whatever" but I can't say I recall that being qualified by ".... provided, of course, the Zs at the end of the new spur is not above the 'maximum Zs' permitted in relation to the circuit's OPD". Do you recall that often (if ever) being added ?

The only thing I have noticed of that one John is "Must be no longer than 2 metres" which perhaps shows a little thinking about it should not be too long but a complete misunderstanding of the Two Metre Rule itself

As for the 8 metre idea, then I have no idea where it originates unless as a means of getting about 3/4 of all of them within sensible limits by the "laws of averages" in some perverse way.

Personally, I have been known to do other that the double ring swap-over I also include test of spurs to get the actual R1 + R2 reading from a spur I suspect might me nearing midpoint and often a Zs too

 

[JohnW2]

What's the point most people haven't got a tester to measure it anyway.

That's true - but it seems a bit odd that people are very often told that what they propose to do would be "fine" (because it's originating at a socket 'on the ring' and is supplying only one socket), even though the length of the proposed spur has often never been mentioned.

Without any means of testing they obviously could (be told how to) calculate how much Zs would be added by their spur cable, but that would not really help if (as is usually going to be the case) they don't know what the Zs was at the origin of the spur.

 

[JohnW2]

The only thing I have noticed of that one John is "Must be no longer than 2 metres" which perhaps shows a little thinking about it should not be too long but a complete misunderstanding of the Two Metre Rule itself

I can't say that I even recall that (in the present context) and, in any event, maybe people should be 'warned' that a spur of any length is theoretically not permissible if the Zs at the proposed origin of that spur is already at (or very close to) the maximum allowed for the circuit.

 

[pete01]

That's true - but it seems a bit odd that people are very often told that what they propose to do would be "fine" (because it's originating at a socket 'on the ring' and is supplying only one socket), even though the length of the proposed spur has often never been mentioned.

Without any means of testing they obviously could (be told how to) calculate how much Zs would be added by their spur cable, but that would not really help if (as is usually going to be the case) they don't know what the Zs was at the origin of the spur.

90% have rcd protection anyway

 

[JohnW2]

As for the 8 metre idea, then I have no idea where it originates ...

If you're talking about the OSG, it's not "8 metres" but, rather 1/8 (12.5%) of the distance from origin of ring to midpoint.

 

[JohnW2]

90% have rcd protection anyway

True - but (in TN installations) the regard that as 'additional protection' (additional to OPD-based ADS). Are you suggesting (the regs obviously don't, at least not yet!) that, in the presence of RCD protection, we can 'forget about' Zs ?

 

[pete01]

True - but (in TN installations) the regard that as 'additional protection' (additional to OPD-based ADS). Are you suggesting (the regs obviously don't, at least not yet!) that, in the presence of RCD protection, we can 'forget about' Zs ?

No I'm not im just saying a question we do ask is "does the circuit have rcd protection" and if yes I wouldn't be losing sleep over the Zs.....maybe thats just me

 

[JohnW2]

No I'm not im just saying a question we do ask is "does the circuit have rcd protection" and if yes I wouldn't be losing sleep over the Zs.....maybe thats just me

Same here, in the sense that I certainly wouldn't lose any sleep over it if it were my own installation. However, we're talking about the regulations, and advice given to others, and that's obviously rather a different matter.