Recessed Lighting: Fire Rated Or Not?

[securespark]

Have looked at the table provided on the esf website.

Trying to work out from the table whether fire-rated down lighters are required in a ground floor room of a two storey detached house.

To me, it's not easy to interpret the table.

What do you think?

 

[EFLImpudence]

Note the word 'guidance'.

The only ground floor mentioned for two story is 'room over basement'.

 

[RF Lighting]

I think it’s poorly worded but first floor is referring to the ceiling between the ground floor and first floor, and then roof ceiling is the ceiling between the first floor and the loft.

Also I believe their guidance to be incorrect. There is no requirement to install fire rated down lighters in a two story dwelling as the entire building is classed as a single fire compartment.

I don’t know about your house but my house has a huge hole in the ceiling between the ground floor and the first floor with a pile of flammable wood shoved in it. It’s where the stairs go through. A 70mm hole with a downlighter in it is going to be fairly insignificant compared to the stairs.

Fire rated down lighters = more profit to the manufacturers.

 

[securespark]

Entirely in agreement. Thank you for your interpretations.

If I can give you a bit of background as to why I posted this thread.

At least 95% of our electrical installation is as it was when we bought the house, barring accessory or luminaire changes, or the odd additional socket.

At some point in the future, we may sell up, and I want to have an EICR and a board change. For two reasons, I don't want to do that myself: I want it to be seen as an impartial job and don't want the hassle of involving LABC.

So, I wanted to find firm documentary evidence I can refer to that my lounge down lighters do not need changing for fire-rated units because I know damn well a large number of sparks doing EICRs will write off non-rated recessed lighting.

The other thing was the end to end CPC reading on a ring final.
I had a new boiler a few years ago and the (NIC) spark complained the reading was too low!
I suggested it could be a parallel path or interconnection, but that was pooh-poohed: what do I know?

 

[Adam_151]

The other thing was the end to end CPC reading on a ring final.
I had a new boiler a few years ago and the (NIC) spark complained the reading was too low!
I suggested it could be a parallel path or interconnection, but that was pooh-poohed: what do I know?

Probably through the boiler he had just connected......

That generally just gets a comment in the notes field of that circuit, "parralell paths to r2 conductor" as a commentry for anyone glancing at the readings and unable to make the jump themselves...

I suppose it would mask a poor/open connection on r2, but it also kind of negates it at the same time too, except perhaps at the one point where the problem was, but as long as you have verified earthing at every point as well, then its likely to be less of a risk than the dismanting required to remove the paralell paths to ensure that you don't have a problem with r2

 

[JohnW2]

So, I wanted to find firm documentary evidence I can refer to that my lounge down lighters do not need changing for fire-rated units because I know damn well a large number of sparks doing EICRs will write off non-rated recessed lighting.

You surely have more than enough knowledge to be able to quiz anyone you are considering appointing to do an EICR about their views/attitudes to such things (including, if appropriate, plastic CUs!) before you decide which one to commission to do the EICR, don't you? That's certainly what I would do in an equivalent situation.

The other thing was the end to end CPC reading on a ring final.
I had a new boiler a few years ago and the (NIC) spark complained the reading was too low!
I suggested it could be a parallel path or interconnection, but that was pooh-poohed: what do I know?

Probably through the boiler he had just connected...... That generally just gets a comment in the notes field of that circuit, "parralell paths to r2 conductor" as a commentry for anyone glancing at the readings and unable to make the jump themselves...

I'm a bit confused here. If it truly were and end-to-end measurement on the CPC, then I can't see how 'parallel paths' (to where? earth?) could result in the measurement being 'too low'. The only thing I can think of that could do that would be a cross-connection between the two sides of the ring (i.e. a 'figure-of-eight' circuit'). What am I missing? Was it perhaps not a true "end-to-end measurement on the CPC"?

Kind Regards, John

 

[RF Lighting]

It’s less common in domestic properties so that may be why the electrician was confused but here’s an example:

Imagine you have a ring circuit supplying 10 outlets.

Outlet number one is an FCU supplying the central heating boiler.

Outlet number eight on the ring is an FCU supplying a waste disposal unit under the kitchen sink.

There is now a path from outlet number one via the CPC to the boiler which connects to the copper water pipework which also connects to the kitchen sink and the CPC of the waste disposal. This bypasses sockets two through seven with the much larger CSA conductor of the pipework , so the end to end reading will be lower than expected.

It happens all the time in commercial and industrial properties with metal clad accessories.

 

[JohnW2]

It’s less common in domestic properties so that may be why the electrician was confused but here’s an example: Imagine you have a ring circuit supplying 10 outlets. ... Outlet number one is an FCU supplying the central heating boiler. ... Outlet number eight on the ring is an FCU supplying a waste disposal unit under the kitchen sink. .........

Thanks - and fair enough.

As you presumably realise, I was thinking of a ring which just had sockets - or, at least, sockets and FCUs, with only one of the FCUs (like the boiler) creating a path from the ring's CPC to pipework/earth/whatever.

As you say, if there are two or more FCUs on the same ring whose CPCs are both/all connected to to the same things (pipework etc.), then that will represent the 'cross-connection' I referred to (turning the CPC into a 'figure-of-eight', rather than a simple ring). However, I would (as you say) imagine that's a pretty unusual situation in a domestic setting.

So, secure ... where is the second FCU (or whatever) on your ring which is resulting in a second connection from ring CPC to some pipework/whatever?

Kind Regards, John

 

[securespark]

After the guy left, I disconnected the boiler from the circuit and tested the CPC end to end: it was still low.

So the interconnection must be elsewhere. But I haven't gone looking for it yet.

 

[JohnW2]

After the guy left, I disconnected the boiler from the circuit and tested the CPC end to end: it was still low. So the interconnection must be elsewhere. But I haven't gone looking for it yet.

Interesting and strange!

How low? (and, if you know them, what are the corresponding end-to-end figures for L and N {or L+N}).

Unless there were literally a 'cross-connection' (i.e. a piece of cable/wire), which seems extremely unlikely, a 'functional cross-connection' would (per the RF scenario) presumably have to be consequence of two 'hard-wired' loads (which, in practice, will presumably mean via FCUs from the ring), both of which were connected to pipework/whatever which were in electrical continuity with one another - something that I would have thought would be pretty unusual in a domestic context. Do you even have two (or more) FCUs connected to that ring (and, if you do, what are they connected to?)?

One other ('left field') thought. I don't suppose some kind soul has, at some point in the past, used two different places on the ring to connect supplementary bonding conductors (bonding two things whose exposed-c-ps are already in continuity with one another), have they?

Kind Regards, John

 

[securespark]

I did test all the conductors E2E.

But it was a while back. I can't remember the figures, just that the CPC was quite a lot lower than I would have expected. Having done a bit of exploring, the circuit is a mix of imperial and metric.

Forgot to say there is one FCU attached to a towel rad on the same circuit, but that was disconnected too.

I will do more digging, but it won't be for a while as everyone is home and using tinterweb.

 

[JohnW2]

I did test all the conductors E2E. But it was a while back. I can't remember the figures, just that the CPC was quite a lot lower than I would have expected. Having done a bit of exploring, the circuit is a mix of imperial and metric. Forgot to say there is one FCU attached to a towel rad on the same circuit, but that was disconnected too. I will do more digging, but it won't be for a while as everyone is home and using tinterweb.

Fair enough.

One last wild thought/question for the night ... I presume that none of the cable is in (earthed) conduit, or even that there is some metal conduit attached to some of the back boxes even if not being used (and than none of the back boxes are attached to metal walls ) ?

Kind Regards, John

 

[securespark]

Sadly nothing like that.

 

[JohnW2]

Sadly nothing like that.

No surprise - I was just scraping the barrel of my mind for further (increasingly unlikely) ideas

To summarise, if there really is 'too low' an end-to-end resistance of the ring's CPC, then you are obviously looking for something which is, at least functionally, a 'cross-connection' between CPCs at two (or more) points on the ring - so you will be looking for at least 'two things'. As I see it, the possibilities that have been considered are:

• a literal 'cross-connection' (wire/cable). That seems very unlikely
• as per RF, the most likely would seem to be that there are two (or more) things to which the ring's CPC is 'hard wired' (probably through an FCU) and which have electrical continuity with one another through pipework/whatever.

The more outlandish 'left field' thoughts that have come into my mind (all much less likely, and most already discounted) include:

• two (or more ) Supplementary Bonding conductors connected to CPC at different points on the ring (and connected to exposed-c-ps that are also otherwise connected to each other)
• metal conduit connecting two or more CPC-connected backboxes (whether the conduit is 'in use' or not)
• two or more CPC-connected backboxes (or metal sockets/accessories) in electrical contact with each other (e.g. metal sockets mounted on metal trunking) and/or structural metal of the building

I might have some even wilder thoughts given a little time! In all cases, you will obviously be looking for at least two things that are connected to the ring's CPC (at different places)

Good luck. Keep us posted!

Kind Regards, John

 

[plugwash]

Another possible cause of a "too low" reading (compared to the expected value for a ring wired in T&E with no paralell paths) would be if parts of the ring were wired in something other than T&E. For example flex has a "full size" CPC and I think NYY-J does too. 3 core SWA with a core used as CPC ends up with a "full-size" CPC and then the Armour on top of that.