Could someone (Adam?) please point me at the relevant slice of the
17th Ed, dealing with the number of circuits needed, both lighing and
ring. I've got a couple of on-sites and Tricker's "Regs in Brief", but
can't find it specifically in any of them.

I'm familiar with the 100m^2 floor area limit on rings, also the
recommendation to separate out the kitchen. However I understand
there's also an aspect for splitting circuits floor by floor, and I
can't find the primary source of this. In particular, how strong is
the requirement to split circuits across the upper two floors of a
three storey house, when the load up there (bedrooms and bathroom) is
fairly low anyway?

Also where does the prohibition on borrowing neutrals across stairway
lighting come from, and what are the implications of this. Is there
(AFAIR) also some requirement on feeding stairways from separate
circuits, so as to maintain some lighting after breakers tripping?

Thanks

On 04/11/10 12:55, Andy Dingley wrote:
Could someone (Adam?) please point me at the relevant slice of the
17th Ed, dealing with the number of circuits needed, both lighing and
ring. I've got a couple of on-sites and Tricker's "Regs in Brief", but
can't find it specifically in any of them.

I'm familiar with the 100m^2 floor area limit on rings, also the
recommendation to separate out the kitchen. However I understand
there's also an aspect for splitting circuits floor by floor, and I
can't find the primary source of this.

Not that I'm aware of. A sensible design will try to:

Make circuits reasonably logical

Ensure the circuit layout is appropriate for the expected loads

(not exclusive)

My dormer bungalow has the following design:

Ground floor hall + right side bedrooms, 32A ring 1 (tight geographically)

Kitchen - 32A compact ring serving a high load worktop (2 machines,
kettle, microwave, toaster etc

Rest of kitchen + 3rd bedroom behind it (both are on the left of the
hall) - 32A ring 3. Most of the kitchen sockets on that ring are
expected to be low load (laptop at dining table, floor lamps, that sort
of thing.

Whilst one could argue it would be logical to have all bedrooms on on
ring, it was more logical for cable run reasons to have bedroom 3 and
the kitchen on one.

Lighting is done on a front/back split - ie 10A for back upstairs and
downstairs;

10A for front upstairs and downstairs.

As all the wiring goes through under the upper level floorboards, this
is logical (though unconventional). It also has the advantage that a
failure in one circuit does not black out an entire floor.


In particular, how strong is
the requirement to split circuits across the upper two floors of a
three storey house, when the load up there (bedrooms and bathroom) is
fairly low anyway?

Allow for the fact that if the heating fails, you might have an electric
heater or two going up there. But as you say, generally low average load.

Also where does the prohibition on borrowing neutrals across stairway
lighting come from, and what are the implications of this.

You don't do it, ever - although you do come across old bodges.

If you mix the neutrals between circuit A and B, then you turn off the
MCB on A to work on it, you disconnect A's neutral at the CU only to
find it is now at 240V via some light on circuit B that used A's neutral.

Bad - very bad...

Is there
(AFAIR) also some requirement on feeding stairways from separate
circuits, so as to maintain some lighting after breakers tripping?

No requirement (at least domestically) - but if you are worried, you can
install a battery backed emergency light.

Cheers

Tim

--
Tim Watts

Kitchen is generally made a separate 32A ring.
If the kitchen shared a ring with the house all the appliances may be
at one end of the ring, thereby creating an imbalanced ring by design
(one leg carrying 24A, other leg carrying 8A). You can use a 32A
radial via 4mm FTE if wiring space is restricted, the length limit of
32A radials is avoided by use of an RCD. A radial wired in 4mm FTE has
a single 1.5mm CPC, creating a higher EFLI than that obtainable by a
ring wired in twin 2.5mm FTE with two 1.5mm CPCs, hence the length
limitation where an RCD is not used.

Borrowed neutrals are not permitted.
To do so breaches regs from decades back, but also creates a serious
electrocution hazard: you test dead only to remove the wire and find
it becomes live in your hand because it is still supplied from a
circuit breaker you did not turn off.

Hall lighting.
If redoing the hall lighting, use 3C+E, consider emergency lights
(small LED units exist), perhaps put the hall on its own circuit
although there is no requirement or current recommendation to do so. I
know of one EE who has his hall lights on their own RCD protected
circuit and supplied by a BS1362 cartridge fuse - his stairs would
trouble a mountaineer and continue down past the ground floor to a
limestone floor cellar :-)

Smokes.
Smoke alarms are a pertinent matter when you have 3 stories, wire in 3C
+E and check any existing have not been wired in 2C+E with the earth
used as the interconnect - absolutely non-compliant like borrowed
neutrals.

If you are tearing the place apart, and this is a 3 storey house,
leave in some extra 20mm or 25mm floor to ceiling conduit for any
future expansion - telephone, network, satellite, power, X10,
whatever. We seem to be creating more wires rather than reducing them
- copper association's the world over will be proud :-)

Fix borrowed neutral on hall lights.
If you merely want to rectify a borrowed neutral, you can buy 6181Y in
1.0mm & 1.5mm online cheaply (sheathed singles), or 6241Y likewise
(sheathed single with CPC) which can be used to solve the problem with
a little work. For example if the hall borrows the live off the
downstairs circuit and uses the upstairs for the neutral, you run a
live from the upstairs light switch (permanent live) to the hall light
switch which is re-run in 3C+E. The upstairs light switch is usually
pretty close to the hall light switch, so it can be very easy and in
zone (horizontal along wall from upstairs light switch to hall light
switch on the landing).
Alternatively you can sometimes rewire with occupancy sensors (like a
PIR) thereby eliminating conventional light switches, no idea if there
is a restriction on their use in a hall. Alternatively go french with
a remote relay box and converting the existing switches to push-type
for signalling. Alternatively using a wireless switch technology like
MK Echo or something-similar-by-GET (MK is battery-less based on
Piezo, others use batteries).

Andy Dingley wrote:
Could someone (Adam?) please point me at the relevant slice of the
17th Ed, dealing with the number of circuits needed, both lighing and
ring. I've got a couple of on-sites and Tricker's "Regs in Brief", but
can't find it specifically in any of them.

I'm familiar with the 100m^2 floor area limit on rings, also the
recommendation to separate out the kitchen. However I understand
there's also an aspect for splitting circuits floor by floor, and I
can't find the primary source of this. In particular, how strong is
the requirement to split circuits across the upper two floors of a
three storey house, when the load up there (bedrooms and bathroom) is
fairly low anyway?

There are no rules about having upstairs/downstairs circuits. It is the
traditional way of wiring the circuits but it is not a rule. It would make
far more sense in many houses that have concrete floors and a CU in the
kitchen to have the kitchen on it's own ring and the rest of the house on
another.

It is worth noting that the 17th edn OSG gives a maximum cable length due to
voltage drop with a 32A type B MCB of 106m for a ring (limited to 46m for a
non RCD TN-S supply due to the earth loop impedance).

Also where does the prohibition on borrowing neutrals across stairway
lighting come from, and what are the implications of this. Is there
(AFAIR) also some requirement on feeding stairways from separate
circuits, so as to maintain some lighting after breakers tripping?

314-01-04 each final circiut shall be connected to a seperate way in a
distribution board. The wiring of each final circuit shall be electrically
seperate from that of every other final circuit.

So if the landing light takes its live from the downstairs lighting MCB but
has the neutral connected to the upstairs lighting neutral then turning the
upstairs lighting MCB off to change a bedroom light fitting will not
guarantee that the bedroom light is not live.

Again there is no requirment to have the landing light on its own circuit.
Most houses still have an upstairs downstairs lighting setup and most people
manage quite well with that set up.

Cheers
--
Adam

On Nov 4, 2:52*pm, "js.b1" wrote:

Borrowed neutrals are not permitted.

Yes, but where's the section in the 17th Ed that say this? For it
probably says a bit more than "Borrowing a neutral, it's bad like
stealing" and I'd like to check the details too.

it becomes live in your hand because it is still supplied from a
circuit breaker you did not turn off.

Circuit breaker? Lighting from more than one circuit? Luxuries!
8-) All of my lighting is currently supplied through one twist of
fusewire. Except for an outside light, which (by proximity) appears to
be piggy-backed on an extractor fan, which is supplied by something in
the kitchen. Something with suspiciously fat cable. So yes, that's an
outside light, wired in ancient damp string, with no RCD and a 32A
cooker fuse...

know of one EE who has his hall lights on their own RCD protected
circuit and supplied by a BS1362 cartridge fuse

That's rather what I _don't_ want, as I'd prefer to avoid single point
failures on the stairs. I don't need _much_ light on the stairs -
switching on any of the three is adequate light to navigate them in
extremis. Yet if I start splitting the lighting by floors, I'd best
check the full rules against borrowing beforehand.


Smoke alarms are a pertinent matter when you have 3 stories,
Yes, I should probably think about that too. We're just a million
separate battery ones at present (currently having the Autumn chorus
of cold, dead PP3s)

check any existing have not been wired in 2C+Echeck any existing have not been wired in 2C+E

2C+E ? You mean something that isn't brittle black rubber? 8-)


If you are tearing the place apart, and this is a 3 storey house,
leave in some extra 20mm or 25mm floor to ceiling conduit for any
future expansion - telephone, network, satellite, power, X10,

If I could think of somewhere to run it, I would. There just isn't any
obvious straight path on which to do it. I might yet build a boxed
conduit from top to bottom along the only solid wall. The other walls
are lath and plaster and don't line up between floors.


The upstairs light switch is usually
pretty close to the hall light switch,

It's not. It's on a quite separate wooden pattress. I also have no
idea where most of the 1st floor cables run, except for those that run
alongside a heating pipe (toasty!) The only vaguely new cable in the
building, and probably the worst routed.

Andy Dingley wrote:
Could someone (Adam?) please point me at the relevant slice of the
17th Ed, dealing with the number of circuits needed, both lighing and
ring. I've got a couple of on-sites and Tricker's "Regs in Brief", but
can't find it specifically in any of them.

Section 314 of the 17th edition

"Division of Installation"

314.1

Every installation shall be divided into circuits as necessary, to:
(i) avoid hazards and minimize inconvenience in the event of a fault.
(ii) facilitate safe inspection, testing and maintenance (see also sec 537).
(iii) take account of danger that may arise from the failure of a single
circuit such as a lighting circuit.
(iv) reduce the possibility of unwanted tripping of RCDs due to
excessive protective circuit conductor currents produced by equipment in
normal operation.
(v) mitigate the effects of EMI.
(vi) prevent the indirect energising of a circuit intended to be installed.

314.2

Separate circuits shall be provided for parts of the installation which
need to be separately controlled, in such a way that those circuits are
not affected by the failure of other circuits, and due account shall be
taken of the consequences of the failure of any single protective device.

314.3

The number of final circuits required, and the number of points supplied
by any final circuit shall be such as to facilitate compliance with
chapter 43 for overcurrent protection, sec 537 for isolation and
switching and chapter 52 as regards current-carrying capabilities of
conductors.

314.4

Where an installation comprises more than one final circuit, each final
circuits shall be connected to a separate way in a distribution board.
The wiring of each final circuit shall be electrically separate from
that of every other final circuit, so as to prevent indirect energising
of a final circuit intended to be isolated.

====================

ECA - ADEQUATE PROVISION OF ELECTRICAL SOCKET OUTLETS IN THE HOME (PDF)
http://www.uat.eca.co.uk/filelibrary...ad/?FileID=196

Hager 17th edition consumer units guide
http://www.hager.co.uk/service/downl...guide/5632.htm





I'm familiar with the 100m^2 floor area limit on rings,

This is a guidance limit.

17th Appendix 15 1 (iv).

Ring final circuits Reg. 433.1.5

The load current in any part of the circuit should be unlikely to exceed
for long periods the current carrying capacity of the cable (reg 433.1.5
refers). This can generally be achieved by:
(i) locating socket outlets to provide reasonable sharing of the load
around the ring.
(ii) not supplying immersion heaters, comprehensive electric space
heating or loads of a similar profile from the ring circuit.
(iii) connecting cookers, ovens and hobs ith a rated power exceeding 2kW
on their own dedicated radial circuit.
(iv) taking account of the total floor area being served (As a rule of
thumb, a limit of 100m2 has been adopted).


The relevant comments for radial circuits a

20A overcurrent protective device - Rule of thumb for floor area served
is 50m2.

30A or 32A overcurrent protective device - Rule of thumb for floor area
served is 75m2.

The above information is also summarised in Appendix 8 of the on-site
guide, where the areas served are prescribed. I guess the on-site guide
needs to be amended to make it clear that the areas are rules of thumb
not absolute limits.


also the
recommendation to separate out the kitchen. However I understand
there's also an aspect for splitting circuits floor by floor, and I
can't find the primary source of this.

All good practice and passing compliance with sec 314.

In particular, how strong is
the requirement to split circuits across the upper two floors of a
three storey house, when the load up there (bedrooms and bathroom) is
fairly low anyway?

No requirement at all as long as the cable isn't too long and you are
confident about complying with sec 314 and sec 433.1.5.


Also where does the prohibition on borrowing neutrals across stairway
lighting come from, and what are the implications of this.

Sec 314 doesn't allow it. Section 521.8 specifically prohibits it.


Is there
(AFAIR) also some requirement on feeding stairways from separate
circuits, so as to maintain some lighting after breakers tripping?


Sec 314 applies.

Thanks

HTH.

On 04/11/10 15:11, Andy Dingley wrote:
On Nov 4, 2:52 pm, wrote:

Borrowed neutrals are not permitted.

Yes, but where's the section in the 17th Ed that say this? For it
probably says a bit more than "Borrowing a neutral, it's bad like
stealing" and I'd like to check the details too.

I would guess it is implied from some rule that says different circuits
shall be electrically separate, except for supplimental bonding, or
something like that?


--
Tim Watts

On 04/11/10 15:23, Dave Osborne wrote:

314.4

Where an installation comprises more than one final circuit, each final
circuits shall be connected to a separate way in a distribution board.
The wiring of each final circuit shall be electrically separate from
that of every other final circuit, so as to prevent indirect energising
of a final circuit intended to be isolated.

That's the one! (re my earlier reply).

--
Tim Watts

On Nov 4, 3:07*pm, "ARWadsworth"
wrote:

There are no rules about having upstairs/downstairs circuits. It is the
traditional way of wiring the circuits but it is not a rule.

Thanks

have the kitchen on it's own ring and the rest of the house on another.

I'm mostly doing that for two other reasons: the CU is in the middle
of the house, adjacent to the kitchen. It's just simpler routing.
Also in a couple of years we might demolish the kitchen (that's roof
and half the walls gone, Grand Designs scale demolition!) and easy
separability then will obviously be useful.

It is worth noting that the 17th edn OSG gives a maximum cable length due to
voltage drop with a 32A type B MCB of 106m for a ring

Not a problem here. It's not that big a house. TT (overhead)
incidentally.


Also where does the prohibition on borrowing neutrals across stairway
lighting come from, and what are the implications of this. Is there
(AFAIR) also some requirement on feeding stairways from separate
circuits, so as to maintain some lighting after breakers tripping?

314-01-04 each final circiut shall be connected to a seperate way in a
distribution board. The wiring of each final circuit shall be electrically
seperate from that of every other final circuit.

Ah! One of those great unstated things, implied by a simple statement
that looks like the bleeding obvious and no more.

So, 3 ring circus (top 2, bottom, kitchen) it is then.

Andy Dingley wrote:
On Nov 4, 3:07 pm, "ARWadsworth"
wrote:

There are no rules about having upstairs/downstairs circuits. It is
the traditional way of wiring the circuits but it is not a rule.

Thanks

have the kitchen on it's own ring and the rest of the house on
another.

I'm mostly doing that for two other reasons: the CU is in the middle
of the house, adjacent to the kitchen. It's just simpler routing.
Also in a couple of years we might demolish the kitchen (that's roof
and half the walls gone, Grand Designs scale demolition!) and easy
separability then will obviously be useful.

It is worth noting that the 17th edn OSG gives a maximum cable
length due to voltage drop with a 32A type B MCB of 106m for a ring

Not a problem here. It's not that big a house. TT (overhead)
incidentally.

It actually does not take long to use up a 100m drum of 2.5 T&E even in a
small house when there are lots of sockets downstairs and they have concrete
floors and high ceilings.

--
Adam

Tim Watts wrote:
On 04/11/10 15:23, Dave Osborne wrote:

314.4

Where an installation comprises more than one final circuit, each final
circuits shall be connected to a separate way in a distribution board.
The wiring of each final circuit shall be electrically separate from
that of every other final circuit, so as to prevent indirect energising
of a final circuit intended to be isolated.

That's the one! (re my earlier reply).


Also, 521.8.2

521.8.2 The line and neutral conductors of each final circuit shall be
electrically separate from that of every other final circuit, so as to
prevent the indirect energising of a final circuit intended to be energized.

Dave Osborne wrote:
Tim Watts wrote:
On 04/11/10 15:23, Dave Osborne wrote:

314.4

Where an installation comprises more than one final circuit, each final
circuits shall be connected to a separate way in a distribution board.
The wiring of each final circuit shall be electrically separate from
that of every other final circuit, so as to prevent indirect energising
of a final circuit intended to be isolated.

That's the one! (re my earlier reply).


Also, 521.8.2

521.8.2 The line and neutral conductors of each final circuit shall be
electrically separate from that of every other final circuit, so as to
prevent the indirect energising of a final circuit intended to be
energized.

Bugger. That should read:

The line and neutral conductors of each final circuit shall be
electrically separate from that of every other final circuit, so as to
prevent the indirect energising of a final circuit intended to be
*isolated*.

On Nov 4, 3:11*pm, Andy Dingley wrote:
On Nov 4, 2:52*pm, "js.b1" wrote:
know of one EE who has his hall lights on their own RCD protected
circuit and supplied by a BS1362 cartridge fuse

That's rather what I _don't_ want, as I'd prefer to avoid single point
failures on the stairs. I don't need _much_ light on the stairs -
switching on any of the three is adequate light to navigate them in
extremis. Yet if I start splitting the lighting by floors, I'd best
check the full rules against borrowing beforehand.

I think you are trying to light each floor of the stairs by each floor
level's lighting circuit - so if another floor trips off you still get
some light on the stairs?

That can be done by
a) having push-switches and a relay/timer system like the French use -
so any switch on any floor causes all levels of stair lights to turn
on/off - BUT each floor is on a separate lighting circuit for
"redundancy".
....
b) having stair light switches on each level of stairs just connected
to that floor level's lighting circuit - that means you switch on at
the bottom of each stairs and off at the top, the two switches being 2-
way switched, linked via 3C+E (3-core and earth). On the next level of
stairs you have another switch. That may work with landings at each
stairs sort of like a hotel, or it aesthetically may look a bit odd
with switches next to one another re "left switch for stairs behind
you, right switch to stairs above you" although I dunno it has its
attractions if a relative is being a PITA :-)

I think the simplest solution is Hall lights on dedicated RCBO and Em-
Light in case that circuit fails.

The EE created his own Em-Lights via LED GU10 powered off the burglar
alarm 12V 7Ah battery, triggered by NO relay off all the lighting
circuits & smoke alarm circuit. Relatively cheap and more integrated
than conventional Em-Lights.


2C+E ? *You mean something that isn't brittle black rubber? *8-)

Someone mentioned black mortar, where they used coal dust/ash to
colour mortar and replace sand. That mortar, particularly when wet due
to leaks, can cause TRS cable crossing it to degrade sufficiently to
short out badly (close to the meter it welded the MCB contacts and
took out the BS1362 60A main fuse, along with showering plaster
everywhere). The cable sheath was cracked most noticeably at the
mortar crossings, particularly where there was water penetration
(water crossing to inner leaf by brick return linked wall vent above
and a recent guttering leak).

So whilst TRS may appear ok, it can still go bang in a wall for little
reason.

Andy Dingley wrote:
On Nov 4, 2:52 pm, "js.b1" wrote:

The upstairs light switch is usually
pretty close to the hall light switch,

It's not. It's on a quite separate wooden pattress. I also have no
idea where most of the 1st floor cables run, except for those that run
alongside a heating pipe (toasty!) The only vaguely new cable in the
building, and probably the worst routed.

Of course they could closer if you do a rewire. It is unusual not to have
this type of setup.


One very easy way to see the stairs if one MCB trips is to two way the
hallway lights on all floors (each floor has it's own MCB). So on ground
floor you can switch the hall and 1st floor landing light, on the 1st floor
you can switch all 3 floors hallway/landing lights and on the 2nd floor
there is a switch to do the 1st and 2nd floor landing light.

--
Adam

Andy Dingley wrote:
On Nov 4, 3:07 pm, "ARWadsworth"
wrote:

There are no rules about having upstairs/downstairs circuits. It is
the traditional way of wiring the circuits but it is not a rule.

Thanks

have the kitchen on it's own ring and the rest of the house on
another.

I'm mostly doing that for two other reasons: the CU is in the middle
of the house, adjacent to the kitchen. It's just simpler routing.
Also in a couple of years we might demolish the kitchen (that's roof
and half the walls gone, Grand Designs scale demolition!) and easy
separability then will obviously be useful.

It is worth noting that the 17th edn OSG gives a maximum cable
length due to voltage drop with a 32A type B MCB of 106m for a ring

Not a problem here. It's not that big a house. TT (overhead)
incidentally.


Also where does the prohibition on borrowing neutrals across
stairway lighting come from, and what are the implications of this.
Is there (AFAIR) also some requirement on feeding stairways from
separate circuits, so as to maintain some lighting after breakers
tripping?

314-01-04 each final circiut shall be connected to a seperate way in
a distribution board. The wiring of each final circuit shall be
electrically seperate from that of every other final circuit.

Ah! One of those great unstated things, implied by a simple statement
that looks like the bleeding obvious and no more.

So, 3 ring circus (top 2, bottom, kitchen) it is then.

But with 3 lighting circuits, not due to the loading but because you would
not want two floors of lights to go out when a bulb pops.(You will find that
out when you swap the fusebox for a CU)

--
Adam

On Nov 4, 7:53*pm, "ARWadsworth"
wrote:

One very easy way to see the stairs if one MCB trips is to two way the
hallway lights on all floors (each floor has it's own MCB).

At present there's two-way switching on the upstairs floors, with a
second switch on the floor below (Ground floor hallway is only on the
ground).

We'd also like to have "period" switches, which is turning out to mean
crappy anodised-gold aluminium toggles from TLC. I think they look
awful, not much like toggles and not at all like brass, but it's
what's available, what puts 3 switches onto a single faceplate, and
what doesn't cost £17 a toggle, as the decent real brass ones do.

OTOH, they can only support 2-way but not 3-way switching, as they
don't do a crossover.

Another idea is to use ancient tumbler switches, but just run them at
low voltage with a control box elsewhere.

If I powered each light from the _lower_ floor's supply, not its own
floor, I'd have access to stairway lights on two different circuits
from both the upstairs floors (top floor light is enough to see by).
Only the ground floor then loses the ability to turn the lights on
after a trip (and you can go fix it first).

Tim Watts wrote:
If you mix the neutrals between circuit A and B, then you turn off the
MCB on A to work on it, you disconnect A's neutral at the CU only to
find it is now at 240V via some light on circuit B that used A's neutral.

Bad - very bad...

Yes, very nasty, happened to me in a former council house.
Just managed to stop myself tumbling off the stepladder and
down the stairs.

JGH

jgharston wrote:

Tim Watts wrote:
If you mix the neutrals between circuit A and B, then you turn off the
MCB on A to work on it, you disconnect A's neutral at the CU only to
find it is now at 240V via some light on circuit B that used A's neutral.

Bad - very bad...

Yes, very nasty, happened to me in a former council house.
Just managed to stop myself tumbling off the stepladder and
down the stairs.

I don't want to justify or excuse mixing neutrals, I think it's a
daft thing to do. But nevertheless I'd like to ask: Why is it
considered good practice to disconnect *both* live and neutral?

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.

Ronald Raygun wrote:
jgharston wrote:

Tim Watts wrote:
If you mix the neutrals between circuit A and B, then you turn off
the MCB on A to work on it, you disconnect A's neutral at the CU
only to find it is now at 240V via some light on circuit B that
used A's neutral.

Bad - very bad...

Yes, very nasty, happened to me in a former council house.
Just managed to stop myself tumbling off the stepladder and
down the stairs.

I don't want to justify or excuse mixing neutrals, I think it's a
daft thing to do. But nevertheless I'd like to ask: Why is it
considered good practice to disconnect *both* live and neutral?

No one is suggesting that both live and neutral are disconnected. In fact
with a borrowed neutral this would be dangerous.

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.

http://s428.photobucket.com/albums/q...t=untitled.jpg

shows a borrowed neutral. If you turn off MCB1 and then take the light
fitting down and the light getting it's power from MCB2 is switched on then
the blue cable from the this light will become live.

--
Adam

On Nov 5, 11:19*am, Ronald Raygun
wrote:

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.

If neutrals were truly "permanently bonded together" then it would be
OK.

However in practice they're bonded half-a-house away, down cable that
could be unreliable. There will also be a path to live through the
lightbulbs. So if this other path is to a live that isn't isolated
(i.e. borrowed neutral) and is switched on at the same time, and the
neutral loop impedance climbs to a fraction of that of a cold
lightbulb (and remember, there's a fault somewhere, which is why we're
fiddling) then that can be enough to put a harmful voltage onto a
supposedly "dead" circuit.

It's like driving behind one of those Fords with a poorly earthed rear
light cluster, so that it indicates every time it brakes.

Also, if working on something where you don't _know_ there isn't a
borrowed neutral, isolate all the lighting circuits, don't just pull
fuses until the light turns off.

On 5 Nov,
Ronald Raygun wrote:

I don't want to justify or excuse mixing neutrals, I think it's a
daft thing to do. But nevertheless I'd like to ask: Why is it
considered good practice to disconnect *both* live and neutral?

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.

I fitted my loft lights off the ceiling rose for the landing light(powered
from the downstairs lighting circuit) so that any work in the loft on
upstairs lights would be lit.

When I came to move an upstairs light, as soon as I disconnected the neutrals
--flash-- darkness, a shared neutral, and a jumbled loft to cross. It could
have been a lot worse.

I now am certain I haven't any shared neutrals.

--
B Thumbs
Change lycos to yahoo to reply

On Nov 4, 3:07*pm, "ARWadsworth"
wrote:

There are no rules about having upstairs/downstairs circuits. It is the
traditional way of wiring the circuits but it is not a rule. It would make
far more sense in many houses that have concrete floors and a CU in the
kitchen to have the kitchen on it's own ring and the rest of the house on
another.

It is worth noting that the 17th edn OSG gives a maximum cable length due to
voltage drop with a 32A type B MCB of 106m for a ring (limited to 46m for a
non RCD TN-S supply due to the earth loop impedance).

I should have looked at this kind of thing before buying my house -
the electrician who sold it to me had rewired the place. Power
circuits (on 30A rewireable fuses) were 1 cooker point in kitchen,
with single 13A socket integral, and one ring covering cellar, ground,
first and second floors, with fused spur to garage!
Lighting was a little more sensible, though the switches do seem to
have been put where it's convenient to wire rather than convenient to
use - the only swirch for the hall light is 4m away from the front
door.

ARWadsworth wrote:

Ronald Raygun wrote:
jgharston wrote:

I don't want to justify or excuse mixing neutrals, I think it's a
daft thing to do. But nevertheless I'd like to ask: Why is it
considered good practice to disconnect *both* live and neutral?

No one is suggesting that both live and neutral are disconnected. In fact
with a borrowed neutral this would be dangerous.

I thought they were, and in particular that certain types of circuit breaker
would disconnect both live and neutral. Was that a misunderstanding on my
part, do such devices exist, and if so, why do they disconnect both poles?

In an analogous situation to that described by the diagram to which you
refer below, a tripped or disconnected breaker on ciruit 1 would cause
*both* live and neutral of circuit 1 to become live when the lamp in
circuit 2 is switched on, and that's before you even open it up.

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.


http://s428.photobucket.com/albums/q...t=untitled.jpg

shows a borrowed neutral. If you turn off MCB1 and then take the light
fitting down and the light getting it's power from MCB2 is switched on
then the blue cable from the this light will become live.

OK, I see what you mean. In this case the two blues would no longer be
connected to each other.

On 06/11/10 17:00, Ronald Raygun wrote:

I thought they were, and in particular that certain types of circuit breaker
would disconnect both live and neutral. Was that a misunderstanding on my
part, do such devices exist, and if so, why do they disconnect both poles?

I have some 2 pole RCBOs (Hager) for outside circuits. It is not a
*general* requirement, but I feel it isn't a bad idea. One situation
where it would be a requirement is if that circuit fed a remote TT
installation from a non TT source, eg a workshop. TT must have DP RCD
protection. Normally you'd use a DP Type S 100ma RCD followe by a 30mA
RCD for sockets or anything else that required 30mA RCD protection.

But having a 100mA RCD for one 30mA RCD circuit (like a shed supply)
would be a bit unnecessary though perfectly correct.

In an analogous situation to that described by the diagram to which you
refer below, a tripped or disconnected breaker on ciruit 1 would cause
*both* live and neutral of circuit 1 to become live when the lamp in
circuit 2 is switched on, and that's before you even open it up.

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.


http://s428.photobucket.com/albums/q...t=untitled.jpg

shows a borrowed neutral. If you turn off MCB1 and then take the light
fitting down and the light getting it's power from MCB2 is switched on
then the blue cable from the this light will become live.

OK, I see what you mean. In this case the two blues would no longer be
connected to each other.


They would be if they were connected downstream of the CU, unless I'm
not following you?

Cheers

Tim

--
Tim Watts

Tim Watts wrote:

On 06/11/10 17:00, Ronald Raygun wrote:

It seems to me that if one were only to disconnect live, then this
particular problem would not arise, since all the neutrals would in
effect be permanently bonded together, and never live, nor ever
much different from earth.

http://s428.photobucket.com/albums/q...t=untitled.jpg

shows a borrowed neutral. If you turn off MCB1 and then take the light
fitting down and the light getting it's power from MCB2 is switched on
then the blue cable from the this light will become live.

OK, I see what you mean. In this case the two blues would no longer be
connected to each other.

They would be if they were connected downstream of the CU, unless I'm
not following you?

The diagram shows two circuits emanating from the CU.
Circuit 1 feeds live and nuetral to lamp 1.
Circuit 2 feeds live to lamp 2, but lamp 2 takes its neutral from lamp 1.

The idea is that if you switch off circuit 1 (but not necessarily its
nuetral), in order to do some work on lamp 1, like remove it which
involves disconnecting both neutrals from it (both the one coming from
the CU and the one going to lamp 2), that if you then switch on lamp 2
it will make its neutral (which is now open-circuit) live. Lamp 2 will
of course not light because it has lost its neutral feed.

When you then come to re-fit a replacement for lamp 1, you will encounter
an unexpected live neutral coming form lamp 2.

If the two blues were still connected to each other, they would not become
live, unless the MCB switch at the CU was 2-pole.