'Leapfrog' wiring of a ring in a long workshop. What do you think?

[Mark42]

I'm insulating, and replacing the wiring, in one of my own (metal, agricultural-type) workshops - fitting 13A/16A sockets on each of the horizontal 'top hat' wall rails, 20m long, with each 4m bay having 4-5 sockets.

This will be done in stages over the next year: the insulation and plasterboard takes a long time, and I'll completely finish/paint each bay before moving on to the next, so as to have an always-working installation.

The sockets on each wall will be wired as RFCs in 2.5mm T&E inside the top hat rail, closed with Celotex and plasterboard. It provides an excellent ready-made concealed trunking system.

I'd normally wire a ring socket-to-next-socket and so on, then have one often longer run back to the board from the end.

But this time, to make it easy to stop and start work yet still have a functioning ring (by adding a single short link at the end), I'll 'leapfrog' the ring wiring to every second socket.

I have no idea if this is normal practice! Thinking about it, it does feel better-designed anyway, as nearly every socket-to-socket cable run will be exactly the same length.

Have I missed anything?
[automerge]1591826810[/automerge]
Too many words! This is what I mean:

Obviously this question only applies to an 'industrial' application where there are loads of sockets in one long, obvious, visible length.

 

[telectrix]

leapfrogging as in pic 3 is the best way. alternatively, if you have enough ways in the DBseparate radials in 4mm for each bay, or 1 radial to serve 2 bays.

 

[bill01803]

I take it you are doing the 16A on a separate circuit?

 

[Midwest]

Always used that method in commercial/industrial installations.

 

[cliffed]

theres really no difference it’s a ring,whichever way, adding sockets,just split the 2 ends,when finished join the 2 ends.
Marvellous.

 

[Pretty Mouth]

If both legs of the ring are following the same route then I'd say a 4mm radial would be more suitable.

 

[Mark Wright]

I was going to ask in a situation like this does a Ring actually have any benefits over a Radial ?

 

[Pretty Mouth]

I was going to ask in a situation like this does a Ring actually have any benefits over a Radial ?

Compared to a 32A 4mm radial, if both legs following the same route benefits of a ring might be:

slightly longer circuit length in terms of voltage drop,
easier to terminate 2.5mm conductors
benefit of high integrity earthing if needed

So not much really. IMO a circuit like this is crying out to be installed as a radial

 

[davesparks]

I'm insulating, and replacing the wiring, in one of my own (metal, agricultural-type) workshops - fitting 13A/16A sockets on each of the horizontal 'top hat' wall rails, 20m long, with each 4m bay having 4-5 sockets.

This will be done in stages over the next year: the insulation and plasterboard takes a long time, and I'll completely finish/paint each bay before moving on to the next, so as to have an always-working installation.

The sockets on each wall will be wired as RFCs in 2.5mm T&E inside the top hat rail, closed with Celotex and plasterboard. It provides an excellent ready-made concealed trunking system.

I'd normally wire a ring socket-to-next-socket and so on, then have one often longer run back to the board from the end.

But this time, to make it easy to stop and start work yet still have a functioning ring (by adding a single short link at the end), I'll 'leapfrog' the ring wiring to every second socket.

I have no idea if this is normal practice! Thinking about it, it does feel better-designed anyway, as nearly every socket-to-socket cable run will be exactly the same length.

Have I missed anything?
[automerge]1591826810[/automerge]
Too many words! This is what I mean:View attachment 58761
Obviously this question only applies to an 'industrial' application where there are loads of sockets in one long, obvious, visible length.

The drawing you have labelled as a 'standard ring' is incorrect an any situation, the sockets, and therefore the loads, should be evenly distributed around the circuit.

The third picture, labelled as 'leapfrog' is correct and provides the most even distribution of the sockets around the ring.

 

[littlespark]

It might be worth documenting somewhere how you are going to run the cables... even numbering the back of the boxes. Future fault finding could be confusing if you do option 3, and future spark expects option 1.

Also, can you answer the question in #3.... the 16A sockets?

 

[pirate]

Maybe it's just me, but the 3rd diagram is not leap-frog at all...its just wiggly wiring of many more sockets in a ring, adding more sockets to what was previously a single long run , without sockets, back to the db

 

[Pete999]

I'm insulating, and replacing the wiring, in one of my own (metal, agricultural-type) workshops - fitting 13A/16A sockets on each of the horizontal 'top hat' wall rails, 20m long, with each 4m bay having 4-5 sockets.

This will be done in stages over the next year: the insulation and plasterboard takes a long time, and I'll completely finish/paint each bay before moving on to the next, so as to have an always-working installation.

The sockets on each wall will be wired as RFCs in 2.5mm T&E inside the top hat rail, closed with Celotex and plasterboard. It provides an excellent ready-made concealed trunking system.

I'd normally wire a ring socket-to-next-socket and so on, then have one often longer run back to the board from the end.

But this time, to make it easy to stop and start work yet still have a functioning ring (by adding a single short link at the end), I'll 'leapfrog' the ring wiring to every second socket.

I have no idea if this is normal practice! Thinking about it, it does feel better-designed anyway, as nearly every socket-to-socket cable run will be exactly the same length.

Have I missed anything?
[automerge]1591826810[/automerge]
Too many words! This is what I mean:View attachment 58761
Obviously this question only applies to an 'industrial' application where there are loads of sockets in one long, obvious, visible length.

If the containment is conduit or dedicated trunking then no 3 is the bet way, in conduit it can be done with a single run, DB to final socket location, simple really, did it this way on many occasions, also when installed using 4 core MICC as an example, sadly not taught on quick fit electrician courses, so you can understand the confusion.

 

[Mike Johnson]

In a workshop situation I would be concerned about the loading possible with five individual bays with up to five sockets in each being used at the same time, what is going to be powered in each bay?

 

[telectrix]

In a workshop situation I would be concerned about the loading possible with five individual bays with up to five sockets in each being used at the same time, what is going to be powered in each bay?

that's why i suggested 3 -5 separate radials in 4mm.

 

[Mike Johnson]

Yes think that would be the most sensible way forward IMO.

 

[Intoelectrics]

Maybe it's just me, but the 3rd diagram is not leap-frog at all...its just wiggly wiring of many more sockets in a ring, adding more sockets to what was previously a single long run , without sockets, back to the db

An easy way to imagine it is this, where you have a long wall with sockets all in one row along that wall and the db is at one end of the wall then there will be a short leg from the db to the closest socket then a long leg to the furthest socket (not ideal). To avoid this each leg serves every other socket until they finally meet at the end socket which then becomes the middle of the ring and thus evens out the loads across the circuit.
Does that make sense?
As mentioned on this thread though, radials might be a more desirable design. To avoid confusion for anyone following up to fault find in the future.

 

[Lucien Nunes]

There should be no confusion - fig. 3 is a standard circuit installed in the recommended manner. Any electrician who finds that confusing is in the wrong job. As mentioned in multple posts above, installing all sockets in one leg with a long return leg is bad practice as the centre of an evenly-distributed load is only 1/4 of the way along the total length of cable, not 1/2 way.

We do not know the expected load so cannot comment on how many radials etc might be preferable. If it's to power assorted bench lamps, phone chargers and the occasional power tool, multiple 4mm² radials would be rather a waste of effort. If there are a few heavy loads then a ring may not be the best, but the fact that there are 13A and 16A sockets makes me think the OP has already thought about the distinction between general and tool / equipment loads and designed accordingly.

 

[pirate]

certainly, option 3 provides even spacing around the ring, but it is not leapfrogging...it's just option 2, bunched up
A radial would be prettier...

 

[nicebutdim]

certainly, option 3 provides even spacing around the ring, but it is not leapfrogging...it's just option 2, bunched up
A radial would be prettier...

Look at the difference in length of the outward and return legs of those options. I think the drawing might be deceiving you.

If option 2 sockets were numbered in a clockwise direction, the outward leg of option 3 connects 1,3,5,7 etc, with return leg picking up each socket in between.

 

[Pretty Mouth]

Compared to a 32A 4mm radial, if both legs following the same route benefits of a ring might be:

slightly longer circuit length in terms of voltage drop,
easier to terminate 2.5mm conductors
benefit of high integrity earthing if needed

So not much really. IMO a circuit like this is crying out to be installed as a radial

Have just thought of another benefit a ring installed in this manner might have over a 32A 4mm radial: a ring would have more flexibility regarding its method of installation.

 

[Pete999]

Have just thought of another benefit a ring installed in this manner might have over a 32A 4mm radial: a ring would have more flexibility regarding its method of installation.

Go on. ? ? ?

 

[Pretty Mouth]

Go on. ? ? ?

A 32A 4mm² radial could only be installed ref method C if T&E, or if singles in conduit/trunking method B or C. Other ref methods would bring the cable CCC below 32A.

On the other hand a 32A 2.5mm² ring can be installed ref methods A, B, C, 100, 102 if T&E, or A, B, and C if singles in conduit/trunking.

But I think you already know this

 

[Pete999]

A 32A 4mm² radial could only be installed ref method C if T&E, or if singles in conduit/trunking method B or C. Other ref methods would bring the cable CCC below 32A.

On the other hand a 32A 2.5mm² ring can be installed ref methods A, B, C, 100, 102 if T&E, or A, B, and C if singles in conduit/trunking.

But I think you already know this

Was just trying to understand yor reasoning PM that's all

 

[Simon47]

... fitting 13A/16A sockets on each of the horizontal 'top hat' wall rails ...

The rails are called purlins, and you can buy plastic covers to go over cables in them. No idea where you can buy the covers from, but I've seen them in use.

 

[Mark42]

Crikey this has all kicked off while I've been away
Thanks for your interest. There are many valid questions to which I'll reply properly tomorrow.

To make it all a bit clearer, here's a picture of the existing unfinished workshop.

I'll be using the central 'top hat' purlin as a cable way. The DB will be at the far end. There will be a variety of sockets and other accessories in one straight line on each long wall.

Here's a picture of a corner of my other, finished, workshop which has exactly the same structure as that above. Note how the accessories follow the line of the top hat rails.

The plan is to convert uninsulated agricultural stores into pleasant office-like environments which I can rent out.
It's amazing what a bit of Celotex and plasterboard, some sticky foam and Tek screws can do.

 

[Simon47]

Had another thought ...
How about doing the sockets in groups by bay ? The the ring would go CU - bay 1 - bay 3 - bay 5 - bay 4 - bay 2 - CU. At least then all the sockets in one bar will be on one leg of the ring so less confusion, but you avoid the "long leg" issue.
As you work on a bay, you just need to leave two tails long enought to reach their next sockets and join then in a JB.
Another option apart from a ring would be two raduals in 2.5mm on a 25A breaker (assuming impedances allow, and check cable capacity vs insrallation method ). Link the CPCs at the far end and you can have high integrity earthing and it will improve the EFLI. If this is intended to be office spave then you might want to consider HI earthing and more circuits. A large quantity of IT on one circuit can be problematic in a number of ways. At one past job, we could not power up one circuit unkess we turned the copier off first - that was due to to total inrush current of all the computers, with the copier tipping it over the edge for the B32 breaker.
Thinking a bit more, you coukd do one 25A radial/bay, and run a 4mm earth (if it's mechanically protected) to pick up the far ends of each circuit's CPC. Gives you segregation (c.f. all eggs in one basket) of loads, and in fact a much higher total load capacity.

 

[DefyG]

Leapfrogging! - pretty standard practice, however;
I think if you are renting out individual office space, it might be wise to consider them as individual circuits as the inconvenience of one office 'tripping off' affecting another office might not be greatly appreciated.

Apart from some of the other issues listed by others, a horizontal purlin rail rail makes a great vermin run and pvc t&e might not be the best choice!

 

[Mike Johnson]

As it's going to be office space presumably with different companies in each bay, it reinforce's my and others opinion that you ought to run separate radials to each bay with IMO RCD protection for each bay and a clean earth installation.

I have come across this sort of rented office space before and most of them have a reception with telephone operator and a meeting room, this gives the possibility of shared office's by different companies, more use and revenue, the possibilities are endless.

 

[Lucien Nunes]

Yes, if the bays are used by different organisations I would make them electrically independent so that one user with a faulty appliance cannot disrupt another user's work. This was not clear from the original description: '...replacing the wiring, in one of my own (metal, agricultural-type) workshops.'

 

[Mark42]

Good morning
Wow this has gone off an various tangents! Sorry for the silence - I've been outside digging holes, as one does.
This building is not for multiple occupancy. If it were, there's no way I'd 'share' circuits, that would be terrible design.
It's not that big - just one open 'shed' 20m x 8m, built in steel with standard 4m bays, to be used for rental to engineering companies who use it for specialist research projects. None are particularly power-hungry: a few computers, printers, the occasional big X-ray machine, data acquisition kit, and a few power tools. All under my supervision.
It's not like I'm going to let the buggers plug in a dc Turboencabulator.
The building is remote and has only a 40A per phase supply. I only want loads of sockets to give flexibility and discourage extension leads. Looks pro too
I'll also put in a few 16A 3 phase sockets and supply distro boxes to give a choice of phase in use, to balance the load.

 

[Lucien Nunes]

If the distribution circuit is protected at 40A you might want to avoid 32A final circuits and go for 20A radials instead.

This apparent contradiction was somewhat confusing as to the application:

The plan is to convert uninsulated agricultural stores into pleasant office-like environments which I can rent out.

This building is not for multiple occupancy.

 

[Mark42]

If the distribution circuit is protected at 40A you might want to avoid 32A final circuits and go for 20A radials instead.
This apparent contradiction was somewhat confusing as to the application:

Agreed. That was badly-written at 01.00.

There are three buildings. Each forming one 'environment.'

I never had any intention of protecting at 32A, for, as you correctly surmise, discrimination up-stream.
I'm going to protect these 13A socket rings at 25A, or even 20A. The sole reason for using a ring was to avoid the hassle of terminating 4mm, and to give high-integrity earthing.

My original question was about only what I used the made-up term 'leapfrogging' for, ie the 1-3-5-7-9 / 2-4-6-8-10 wiring technique!
[automerge]1592222049[/automerge]
Here's a pic of the first bay in 'Building B' which I converted this Feb.

The position of all the accessories is directly in front of the middle purlin. Once finished it's sealed both ends in each bay to keep vermin out. (I hope!)

 

[Simon47]

I'm going to protect these 13A socket rings at 25A, or even 20A. The sole reason for using a ring was to avoid the hassle of terminating 4mm, and to give high-integrity earthing.

You don't (or at least, probably do not) need 4mm for a 25A radial - even less so for a 20A radial. You can do HI earthing just by running a separate earth to the far end of the radial.

 

[Pete999]

You don't (or at least, probably do not) need 4mm for a 25A radial - even less so for a 20A radial. You can do HI earthing just by running a separate earth to the far end of the radial.

And terminating the cpcs in separate terminals in the socket outlets, and terminating the cpcs (if it's an RFC in separate terminals in the CU)

 

[davesparks]

I never had any intention of protecting at 32A, for, as you correctly surmise, discrimination up-stream.
I'm going to protect these 13A socket rings at 25A, or even 20A. The sole reason for using a ring was to avoid the hassle of terminating 4mm, and to give high-integrity earthing.

Ah but you said ring, and the only standard ring circuit in the book is a 32A ring circuit, a 25A or 20A ring circuit is non-standard and for most installations quite pointless so we all assume that a ring final circuit will be a 32A circuit.

 

[DPG]

Make sure you check the power requirements for the 'big x ray machine'

 

[Mark42]

Make sure you check the power requirements for the 'big x ray machine'

Stuff all. It’s a big airport machine but has only a 13A plug.

 

[DPG]

Stuff all. It’s a big airport machine but has only a 13A plug.

Ah right, thought it was a medical one.

 

[Mark Wright]

If you did put the Ring on a 20A or 25A seems you'd have the benefits of a Ring and Radial together in conjunction with the Leapfrog wiring method - thinking about it I can't see any drawbacks?

and again if you did use 4mm I say the same?

Sorry I skimmed the thread - this is fixed load stuff? Seems quite sensible?