I don't like hijacking threads so these questions aren't immediately
related to the question being asked about changing a CU.

I have TT earthing. I upgraded the CU to a split unit with MCB's
about 2 years ago. I didn't realise at that moment in time the
significance of TT earthing and its relationship with RCB's, though I
did so shortly afterwards when reconfiguring the wiring to my sheds.
I see now from the DIY Wiki that an RCD is mandated for the lighting
circuits also - this does make sense, but how am I going to implement
it ? I have two lighting circuits which are at the moment on the non-
RCD side of the CU fed via 6A MCB's. Clearly it cannot just be a case
of transferring these to the RCD side of the CU.

The second question arose from the discussion on changing the CU where
John Rumm was talking about nuisance trips from RCD's. I suffer from
nuisance trips of the MCB's when bulbs fail. I gather this is not
uncommon but do wonder what the recommendation is for resolving it ?
If I and other correspondents to this 'column' suffer from these
trips, why doesn't everyone ? The house was rewired 20 years ago.

Thanks

Rob

robgraham wrote:
I don't like hijacking threads so these questions aren't immediately
related to the question being asked about changing a CU.

I have TT earthing. I upgraded the CU to a split unit with MCB's
about 2 years ago. I didn't realise at that moment in time the
significance of TT earthing and its relationship with RCB's, though I
did so shortly afterwards when reconfiguring the wiring to my sheds.
I see now from the DIY Wiki that an RCD is mandated for the lighting
circuits also - this does make sense, but how am I going to implement
it ? I have two lighting circuits which are at the moment on the non-
RCD side of the CU fed via 6A MCB's. Clearly it cannot just be a case
of transferring these to the RCD side of the CU.


It is usual with a TT installation to have a 100mA RCD covering the
whole consumer unit and a 30mA RCD covering the socket outlets, etc.
The 100mA RCD should be time-delayed to give proper discrimination
between the two RCD's.

The second question arose from the discussion on changing the CU where
John Rumm was talking about nuisance trips from RCD's. I suffer from
nuisance trips of the MCB's when bulbs fail. I gather this is not
uncommon but do wonder what the recommendation is for resolving it ?
If I and other correspondents to this 'column' suffer from these
trips, why doesn't everyone ? The house was rewired 20 years ago.

Thanks

Rob


Well, MCB's come in (basically) three types Type B, Type C and Type D
with decreasing sensitivity on the magnetic part of the trip
characteristic as you go from B to D. Since blowing light bulbs cause
surge currents that are detected and cleared by the magnetic part of the
MCB, changing your MCB's from type B (usually deployed in domestic
installations) to type C or D will reduce or eliminate nuisance
tripping, arguably at the expense of some protection.

The reason why some people experience nuisance tripping and some don't
is due to a number of factors, including (but probably not limited to
the following):

1. The make and model of the MCB
2. The ambient operating temperature of the MCB
3. The cross sectional area of the lighting cable (1.5mm2 is more
likely to be a problem than 1.0mm2)
4. The length of lighting cable between the consumer unit and the light
fitting (The shorter the cable the more likely to be a problem)
5. The make of the light bulb

Also, MCBs have an ultimate breaking capacity. This figure will be
printed on the MCB as a number 3000, 6000, 10000 closely surrounded by a
rectangle. This number equates to the maximum fault current in Amps.
that the breaker can safely interrupt. Most consumer unit MCBs are
rated at 6000A (except the old Wylex plug-in replacement jobs, which are
rated at 3000A). I suspect that industrial-quality MCB's (e.g. Merlin
Gerin C60, ABB S270, etc) which are rated at 10,000A are less prone to
this sort of nuisance tripping, but don't quote me on this - my evidence
is purely anecdotal.

HTH

robgraham wrote:
I don't like hijacking threads so these questions aren't immediately
related to the question being asked about changing a CU.

I have TT earthing. I upgraded the CU to a split unit with MCB's
about 2 years ago. I didn't realise at that moment in time the
significance of TT earthing and its relationship with RCB's, though I
did so shortly afterwards when reconfiguring the wiring to my sheds.
I see now from the DIY Wiki that an RCD is mandated for the lighting
circuits also - this does make sense, but how am I going to implement
it ? I have two lighting circuits which are at the moment on the non-
RCD side of the CU fed via 6A MCB's. Clearly it cannot just be a case
of transferring these to the RCD side of the CU.

The second question arose from the discussion on changing the CU where
John Rumm was talking about nuisance trips from RCD's. I suffer from
nuisance trips of the MCB's when bulbs fail. I gather this is not
uncommon but do wonder what the recommendation is for resolving it ?
If I and other correspondents to this 'column' suffer from these
trips, why doesn't everyone ? The house was rewired 20 years ago.

Thanks

Rob

Following a thread here a few years ago on tripping when a lamp fails I
changed my B type MCBs for C type. There was no appreciable change in
the tripping pattern.

Malcolm

Malcolm Race wrote:

Following a thread here a few years ago on tripping when a lamp fails I
changed my B type MCBs for C type. There was no appreciable change in
the tripping pattern.

Malcolm

Yes, turns out the arc currents in failing filament lamps is 60A
upwards,
often significantly upwards, hence the type C also trips. I dont
know if type D would help, not got curves to hand. Fuses are a
simpler solution though.


NT

Rumble wrote:
robgraham wrote:

I don't like hijacking threads so these questions aren't immediately
related to the question being asked about changing a CU.

I have TT earthing. I upgraded the CU to a split unit with MCB's
about 2 years ago. I didn't realise at that moment in time the
significance of TT earthing and its relationship with RCB's, though I
did so shortly afterwards when reconfiguring the wiring to my sheds.
I see now from the DIY Wiki that an RCD is mandated for the lighting
circuits also - this does make sense, but how am I going to implement
it ? I have two lighting circuits which are at the moment on the non-
RCD side of the CU fed via 6A MCB's. Clearly it cannot just be a case
of transferring these to the RCD side of the CU.


It is usual with a TT installation to have a 100mA RCD covering the
whole consumer unit and a 30mA RCD covering the socket outlets, etc.
The 100mA RCD should be time-delayed to give proper discrimination
between the two RCD's.

The second question arose from the discussion on changing the CU where
John Rumm was talking about nuisance trips from RCD's. I suffer from
nuisance trips of the MCB's when bulbs fail. I gather this is not
uncommon but do wonder what the recommendation is for resolving it ?
If I and other correspondents to this 'column' suffer from these
trips, why doesn't everyone ? The house was rewired 20 years ago.

Thanks

Rob


Well, MCB's come in (basically) three types Type B, Type C and Type D
with decreasing sensitivity on the magnetic part of the trip
characteristic as you go from B to D. Since blowing light bulbs cause
surge currents that are detected and cleared by the magnetic part of the
MCB, changing your MCB's from type B (usually deployed in domestic
installations) to type C or D will reduce or eliminate nuisance
tripping, arguably at the expense of some protection.

The reason why some people experience nuisance tripping and some don't
is due to a number of factors, including (but probably not limited to
the following):

1. The make and model of the MCB
2. The ambient operating temperature of the MCB
3. The cross sectional area of the lighting cable (1.5mm2 is more
likely to be a problem than 1.0mm2)
4. The length of lighting cable between the consumer unit and the light
fitting (The shorter the cable the more likely to be a problem)
5. The make of the light bulb

Also, MCBs have an ultimate breaking capacity. This figure will be
printed on the MCB as a number 3000, 6000, 10000 closely surrounded by a
rectangle. This number equates to the maximum fault current in Amps.
that the breaker can safely interrupt. Most consumer unit MCBs are
rated at 6000A (except the old Wylex plug-in replacement jobs, which are
rated at 3000A). I suspect that industrial-quality MCB's (e.g. Merlin
Gerin C60, ABB S270, etc) which are rated at 10,000A are less prone to
this sort of nuisance tripping, but don't quote me on this - my evidence
is purely anecdotal.

HTH

A small form factor bulb such as a 2" mains halogen is more likely
to take out an mcb than say a 1' long filament tubular bulb. CFLs
dont do this of course.


NT

wrote:

Yes, turns out the arc currents in failing filament lamps is 60A
upwards,

Type C has a 0.1 trip mechanism at about 60A

often significantly upwards, hence the type C also trips. I dont
know if type D would help, not got curves to hand. Fuses are a

A 6A type D brings the current up to 120A at a cost of halving the
maximum permitted fault loop impedance to 2 ohms.

simpler solution though.

These can be taken out as well. Much depends on other details of the
installation (supply impedance, cable lengths and size etc).

Personally I don't have problems with a type C, but that is not any
guarantee the same will apply elsewhere.



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

robgraham wrote:
I don't like hijacking threads so these questions aren't immediately
related to the question being asked about changing a CU.

I have TT earthing. I upgraded the CU to a split unit with MCB's
about 2 years ago. I didn't realise at that moment in time the
significance of TT earthing and its relationship with RCB's, though I
did so shortly afterwards when reconfiguring the wiring to my sheds.
I see now from the DIY Wiki that an RCD is mandated for the lighting
circuits also - this does make sense, but how am I going to implement
it ? I have two lighting circuits which are at the moment on the non-
RCD side of the CU fed via 6A MCB's. Clearly it cannot just be a case
of transferring these to the RCD side of the CU.

You could replace the main incomer switch with a MK 6400S 100mA time
delayed RCD. (assuming the maximum current rating is adequate).

Alternatively install a second CU with a 100mA ordinary RCD as the main
incomer, and put the non socket circuits on that. Then remove the split
from the other one, and user that for the sockets with the RCD moved to
the incomer position.

The second question arose from the discussion on changing the CU where
John Rumm was talking about nuisance trips from RCD's. I suffer from
nuisance trips of the MCB's when bulbs fail. I gather this is not
uncommon but do wonder what the recommendation is for resolving it ?
If I and other correspondents to this 'column' suffer from these
trips, why doesn't everyone ? The house was rewired 20 years ago.

As Rumble said, its complicated! ;-) Type Cs may help, alternatively try
a cartridge fuse holder in place of the MCB (Hager do one that fits a CU
in the same form factor as a MCB).

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

In article . com,
wrote:
Yes, turns out the arc currents in failing filament lamps is 60A
upwards,
often significantly upwards, hence the type C also trips. I dont
know if type D would help, not got curves to hand. Fuses are a
simpler solution though.

Not a total one - I've had a fuse blow when a bulb failed.

--
*I started out with nothing, and I still have most of it*

Dave Plowman London SW
To e-mail, change noise into sound.

John Rumm wrote:
wrote:

Yes, turns out the arc currents in failing filament lamps is 60A
upwards,

Type C has a 0.1 trip mechanism at about 60A

often significantly upwards, hence the type C also trips. I dont
know if type D would help, not got curves to hand. Fuses are a

A 6A type D brings the current up to 120A at a cost of halving the
maximum permitted fault loop impedance to 2 ohms.

simpler solution though.

These can be taken out as well. Much depends on other details of the
installation (supply impedance, cable lengths and size etc).

Personally I don't have problems with a type C, but that is not any
guarantee the same will apply elsewhere.


I guess type D would be the best all round (for lighting). You've got
the nearest
to fuse characteristics, easy reset, and no abuse potential.

Now, type D RCBOs?


NT