Just been checking some wiring done in the era of a previous owner of
the property. A ring main, where each double socket had one of the
cable earths connected to the back box, the other to the earth
terminal on the socket, but no box to socket wire, so ring earth
continuity relied on all the back box screws. (now re-wired). Then
two outside MK waterproof sockets, each a radial off the consumer unit
rcd protected side, with the earth connected to the terminal on the
plastic housing but not to the socket, so no earth at all!

Just as well I checked !

When putting in ring mains, I've always individually sleeved the cable
earths, tight twisted them at the bare end then put them in the socket
connector with the backbox earth lead. This way even if the earth
terminal screw worked loose ring earth continuity would be preserved,
but I understand this is frowned on now as it makes breaking the loop
for testing a bit harder.

AWEM

Andrew Mawson wrote:
Just been checking some wiring done in the era of a previous owner of
the property. A ring main, where each double socket had one of the
cable earths connected to the back box, the other to the earth
terminal on the socket, but no box to socket wire, so ring earth
continuity relied on all the back box screws. (now re-wired). Then
two outside MK waterproof sockets, each a radial off the consumer unit
rcd protected side, with the earth connected to the terminal on the
plastic housing but not to the socket, so no earth at all!

Just as well I checked !

I have seen that done several times with plastic patress'! Scarey.


When putting in ring mains, I've always individually sleeved the cable
earths, tight twisted them at the bare end then put them in the socket
connector with the backbox earth lead. This way even if the earth
terminal screw worked loose ring earth continuity would be preserved,
but I understand this is frowned on now as it makes breaking the loop
for testing a bit harder.

One of my pet hates.

--
Adam

On 08/12/10 09:09, ARWadsworth wrote:
Andrew wrote:
Just been checking some wiring done in the era of a previous owner of
the property. A ring main, where each double socket had one of the
cable earths connected to the back box, the other to the earth
terminal on the socket, but no box to socket wire, so ring earth
continuity relied on all the back box screws. (now re-wired). Then
two outside MK waterproof sockets, each a radial off the consumer unit
rcd protected side, with the earth connected to the terminal on the
plastic housing but not to the socket, so no earth at all!

Just as well I checked !

I have seen that done several times with plastic patress'! Scarey.


When putting in ring mains, I've always individually sleeved the cable
earths, tight twisted them at the bare end then put them in the socket
connector with the backbox earth lead. This way even if the earth
terminal screw worked loose ring earth continuity would be preserved,
but I understand this is frowned on now as it makes breaking the loop
for testing a bit harder.

One of my pet hates.


Hi Adam,

Just out of interest, when do you break connections at an accessory in a
socket circuit for testing?

I'm sure there's a good reason, but in my *limited* training for circuit
testing and subsequant practice, I've only found it necessary to break
the circuit at the CU - all socket tests being done from the front.

OK - inspection requires looking inside but not breaking anything.

Lights are different - in that one must bridge out dimmers etc[1]

Just curious.

[1] One of my pet hates - because you are disturbing the circuit under
test and inherently cannot test it fully after re-assembly. I think
*all* dimmers and other sensitive switches should have a hole in the
back side where a shorting pin can be inserted for testing...

Cheers

Tim

--
Tim Watts

In article ,
"Andrew Mawson" writes:
When putting in ring mains, I've always individually sleeved the cable
earths, tight twisted them at the bare end then put them in the socket
connector with the backbox earth lead. This way even if the earth
terminal screw worked loose ring earth continuity would be preserved,
but I understand this is frowned on now as it makes breaking the loop
for testing a bit harder.

Twisting was done with the old multi-strand cable.
Not a good idea with current cable which is either single strand,
or (larger sizes) multi-stand but too big to twist. Good quality
terminals will often take anything from large conductors right
down to flex, but what they won't handle are a mixture of different
sizes.

When I put in a whole ring, I usually set myself a challenge to
get it all installed without cutting through any of the conductors,
and I've managed this a number of times, but not always. Any loose
terminals impact only the socket they're on, and do not accumulate
along the circuit.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

On 08/12/10 11:45, Andrew Gabriel wrote:
In ,
"Andrew writes:
When putting in ring mains, I've always individually sleeved the cable
earths, tight twisted them at the bare end then put them in the socket
connector with the backbox earth lead. This way even if the earth
terminal screw worked loose ring earth continuity would be preserved,
but I understand this is frowned on now as it makes breaking the loop
for testing a bit harder.

Twisting was done with the old multi-strand cable.
Not a good idea with current cable which is either single strand,
or (larger sizes) multi-stand but too big to twist. Good quality
terminals will often take anything from large conductors right
down to flex, but what they won't handle are a mixture of different
sizes.

When I put in a whole ring, I usually set myself a challenge to
get it all installed without cutting through any of the conductors,
and I've managed this a number of times, but not always. Any loose
terminals impact only the socket they're on, and do not accumulate
along the circuit.


Admirable. I don't think I could have done that without a *lot* of
trouble as I have 100% drops from the ceiling downwards.

I do wish manufacturers would adopt captive terminals as standard - the
number of times I have had one wire, especially the smaller CPC
disappear up the side of the screw are uncountable - nearly made me
start twisting the conductors. I did double some over in the end.

That was on GET sockets which are better in this respect, I have seem
some truely dire terminals in my time...

Never had that problem with my Hager RCBOs which are all captive.

--
Tim Watts

Tim Watts wrote:
On 08/12/10 09:09, ARWadsworth wrote:
Andrew wrote:
Just been checking some wiring done in the era of a previous owner
of the property. A ring main, where each double socket had one of
the cable earths connected to the back box, the other to the earth
terminal on the socket, but no box to socket wire, so ring earth
continuity relied on all the back box screws. (now re-wired). Then
two outside MK waterproof sockets, each a radial off the consumer
unit rcd protected side, with the earth connected to the terminal
on the plastic housing but not to the socket, so no earth at all!

Just as well I checked !

I have seen that done several times with plastic patress'! Scarey.


When putting in ring mains, I've always individually sleeved the
cable earths, tight twisted them at the bare end then put them in
the socket connector with the backbox earth lead. This way even if
the earth terminal screw worked loose ring earth continuity would
be preserved, but I understand this is frowned on now as it makes
breaking the loop for testing a bit harder.

One of my pet hates.


Hi Adam,

Just out of interest, when do you break connections at an accessory
in a socket circuit for testing?

I'm sure there's a good reason, but in my *limited* training for
circuit testing and subsequant practice, I've only found it necessary
to break the circuit at the CU - all socket tests being done from the
front.
OK - inspection requires looking inside but not breaking anything.

Lights are different - in that one must bridge out dimmers etc[1]

Just curious.

You really only need to start messing with socket connections for fault
finding.



Tim

--
Adam

On 08/12/10 12:31, ARWadsworth wrote:

Hi Adam,

Just out of interest, when do you break connections at an accessory
in a socket circuit for testing?

I'm sure there's a good reason, but in my *limited* training for
circuit testing and subsequant practice, I've only found it necessary
to break the circuit at the CU - all socket tests being done from the
front.
OK - inspection requires looking inside but not breaking anything.

Lights are different - in that one must bridge out dimmers etc[1]

Just curious.

You really only need to start messing with socket connections for fault
finding.



Tim


Good - just wondered if I missed something...



--
Tim Watts