CU, head and meter enclosures fire resistance

[bernardgreen]

The problem of fallen cables has been known for years,

The Harrow Court fire occurred in a tower block on 2 February 2005 in Stevenage, Hertfordshire, England.

An investigation into the fire found that there was no dry riser outlet on the 14th floor of the building as they were only located on alternate floors. The firefighters connected to the dry riser on the 13th floor, but then to fight the location of the fire, the hose had to pass through firedoors, which were then kept open. This led to the fire to have more oxygen due to the fact that compartmentation was not achieved. The two firefighters died when they got tangled in cable trays which were attached to the ceiling with plastic hooks which melted.


https://www.fbu.org.uk/publication/executive-hs-summary-85-harrow-court

 

[SimonH2]

Looks like I've been conflating reports from different incidents - or at least different reports. That was indeed one of them, and as you say, shows that the problem has been known about for a long time. But I'm sure I've read another one (maybe it was a fuller report from that incident) that talks about temperatures high enough to spall the concrete.

 

[JohnW2]

The problem of fallen cables has been known for years, ..... "The Harrow Court fire occurred in a tower block on 2 February 2005 in Stevenage, Hertfordshire, England. ...." ...

Thanks, but I'm a bit confused as to where the quote you posted came from, since it is not in the Executive Summary to which you linked. In particular, your quote refers to falling cable trays, whereas the nearest to that in the Executive Summary is a reference to the use of plastic trunking for fire alarm cables, which it says is not allowed by BS 5839-1.

It is clear that, as with almost all disasters, the two deaths were the result of the coming together of multiple factors, the absence of any of which could probably have averted the deaths. However, other than for the reference to plastic trunking for fire alarm cables, the Summary seems to say very little about cables or their installation.

It is presumably (probably?) the case that, in the absence of entanglement in cables, the deaths would not have occurred. However, the report says almost nothing about that, focusing on deficiencies in the fire services (including training), local authorities and even, to some extent, the two victims themselves as being contributors to the deaths, the following being a few examples (words in red were italicised in the Summary) ...

2.30. As a conclusion to our Health and Safety Investigation, the FBU believes that the conduct of Hertfordshire Fire & Rescue Service significantly contributed to the deaths of Ff Wornham and Ff Miller in that they failed to comply satisfactorily with the Fire Services Act 2004; the Health and Safety at Work Act 1974; the Management of Health and Safety at Work Regulations 1999; national guidance issued by Her Majesty’s Inspectorate; and failed to act adequately upon relevant Health and Safety Executive (HSE) improvement notice recommendations available to them. Similarly, during the course of the FBU’s Health and Safety Investigation HFRS failed to comply with the Safety Representative and Safety Committee Regulations 1977.

2.31. It is not known whether the smoke alarm in Flat 85 activated or not. However, nobody reported hearing the smoke alarm in Flat 85 sounding at any time. Since it may not have activated and had it done so the occupants may have made their own way to safety, the FBU’s Health & Safety Investigation concludes that Stevenage Borough Council (SBC) may have contributed to the deaths of Ff Miller and FF Wornham in that they failed to undertake a review of the smoke alarm installations in the individual flats at Harrow Court to assess their appropriateness as recommended in BS 5839-6:2004 annex A 4.1 a & b and 5.1 a, b & c.

2.32. ....... the FBU’s Health and Safety Investigation also concludes that SBC may have contributed to Ff Wornham’s death in that they may have failed to ensure their contractor complied with BS 5839-1: 2002;clause 26.2(f) in respect of precluding the use of plastic trunking for securing the electrical cabling of their common area fire alarm system.

2.34. However, this report identifies areas of organisational weakness that the FBU believes led to HFRS (primarily) failing to prevent or adequately reduce the known risks that eventually overcame Ff Wornham and Ff Miller on 2nd February 2005. .....

2.4. ...... This being the case, the FBU believes that establishing the cause of their deaths has less to do with the precise definition of the fire phenomenon that occurred and more to do with those factors that led to Ff Wornham and Ff Miller opening the front door when they did and without the necessary water, resources or personnel they required.

The investigation identified many organisational weaknesses in the development, monitoring and review of Standard Operating Procedures. ............... 2.11. The FBU considers these Standard Operating Procedures produced by Hertfordshire Fire & Rescue Service (HFRS) were inadequately drafted, monitored and reviewed and as a result, were not fit for purpose at the time of the Harrow Court Incident.

No mention anywhere that I can see of any discussion or recommendations in relation to wiring practices, systems or regulations (other than the reference to the fire alarm cables). Although I've obviously only seen the Executive Summary, it would seem improbable that the full report will have dealt with such things if there is no mention at all in the Executive Summary.

Kind Regards, John

 

[frank999]

The link to the Shirley Towers incident is frightening, cables everywhere.

Can anyone confirm the use of nylon plugs when using FR conduit with the metal clips - is there any Fire Test evidence, nylon plugs would surely melt and the fixing fail, I note Fisher make metal Plugs, and Dewalt 'wall dog' fixings.

I have re-contacted with my Fire Risk Assessors, who have advised that providing the CU is current to regulations (ie metal encasement) that will suffice, iand no need for fire rated mounting board.

If the CU is mounted on two battens to allow rear cable entry (cables running on surface of 60 minute stud wall), does the rear hole require any form of fire stopping ... ie with an intumiscent putty pad or the like.

 

[SimonH2]

The link to the Shirley Towers incident is frightening, cables everywhere.

Yes indeed. In hindsight, multiple issues that should have been seen as problems in their own right, and bigger problems collectively.

Can anyone confirm the use of nylon plugs when using FR conduit with the metal clips - is there any Fire Test evidence, nylon plugs would surely melt and the fixing fail

We've had some discussion around that in here.
On the one hand, when the substrate gets hot enough, the plastic will melt. That means the substrate typically getting that hot to a depth of an inch or more - by which time I suspect even the F&RS won't be hanging about.
But even then, unless the screw is under a tensile only load, it will pull to one side and probably stay put. Just try it, drill a hole, put a screw in without a plug, and hang something from it - as in hanging rather than pulling. The screw will jam in the hole and not fall out. If there's a combination load (as would be typical in most cases), it would depend on various factors, but I would suspect that in most cases we're talking about the screw would not completely fall out - "release" the tensile load, then jam when the shear load takes over.
Thinking about it, unless the heat is coming from the other side of the wall, the plastic will melt progressively as the heat penetrates. So the outer end of the screw would be able to move sideways (and jam) while it is still held in tension from deeper in the hole. That's just my postulation - no facts to back it up.
Size of hole relative to screw size (smaller is better) and screw length (longer is better) will have a significant impact - so just using a bit longer screw in some places would make a big difference.
It would be interesting to know if anyone had done any research on this. It would be particularly useful for "upgrades" to existing installations (e.g. mini-trunking). There's a difference in work required between "remove a screw, add a metal cable support, fit longer screw" and "as before but drill out or otherwise remove plastic plug before fitting new metal one".

If the CU is mounted on two battens to allow rear cable entry (cables running on surface of 60 minute stud wall), does the rear hole require any form of fire stopping ... ie with an intumiscent putty pad or the like.

There's never been any requirement for fire-stopping of a CU. The "non-combustible" (whatever than means, we've discussed it considerably) relates to whether the CU case itself burns when heat is applied from an internal fault. There's nothing in the regs about containing any fire within the CU.
If you did want to "fire stop" it, then I'd suggest an intumescent sock of an appropriate size to fill the gap (between case and wall). That way it can be easily removed and replaced when future works are needed.

 

[flameport]

Can anyone confirm the use of nylon plugs when using FR conduit with the metal clips - is there any Fire Test evidence,

Some tests were done by BRE, report here: https://www.bre.co.uk/filelibrary/Fire and Security/FI---Cables-and-support-fixings-experiments.pdf
However that was very limited in scope, and the temperatures used were not particularly high relative to temperatures found in real building fires. Despite those limitations, various fixings did fail.

For conduit, these: https://www.linianclip.co.uk/product/linian-superclip/ are fire resistant and have been tested to multiple standards.

 

[SimonH2]

Some tests were done by BRE, report here: https://www.bre.co.uk/filelibrary/Fire and Security/FI---Cables-and-support-fixings-experiments.pdf

As above, I'd expect plastic plugs to fail when the fixing is in tension - the only mode tested for that report. I'd expect different results if the fixings were into a wall rather than ceiling.

 

[JohnW2]

As above, I'd expect plastic plugs to fail when the fixing is in tension - the only mode tested for that report. I'd expect different results if the fixings were into a wall rather than ceiling.

Indeed, with plugs in a wall, even if the plug totally melted and disappeared, one would presumably need some other forces (other than gravity) at work before the screw would come out.

Kind Regards, John

 

[frank999]

one would presumably need some other forces (other than gravity) at work before the screw would come out
Screw Fixings in ceilings using plastic/nylon fixings then would fail

Thanks for the comments.

My spark has suggested I do not need SPD protection as I am not in a rural area.
A single bus Wylex CU will be used.
The CU will be all RCBO's, as suggested by my spark - Type A, B Curve.

All the habitable rooms will all be on separate radial circuits, rated at 16amp, the kitchen rings will be on 20amp.

The CU will be mounted on Fire Rated chip board with sufficient holes for the cables, the FR board mounted on 2 battens to create a gap down the back and stand off, and allow rear entry via the knockouts, unsure whether any intumiscent putty should go in the hole.

 

[JohnW2]

one would presumably need some other forces (other than gravity) at work before the screw would come out ... Screw Fixings in ceilings using plastic/nylon fixings then would fail

Yes, obviously.

My spark has suggested I do not need SPD protection as I am not in a rural area.

I have to say that, unlike some others, I'm very sceptical about the 'need' for SPD protection for any domestic installation but there is certainly an argument that any such 'need' is likely to be greater in rural areas with lots of overhead supplies/distribution. However, I've been in such an area for getting on for 40 years, and am unaware of anything that has happened in that time that an SPD would have prevented.

A single bus Wylex CU will be used. ... The CU will be all RCBO's, as suggested by my spark - Type A, B Curve.

I imagine that a good few people here will be very happy with all that.

All the habitable rooms will all be on separate radial circuits, rated at 16amp, the kitchen rings will be on 20amp.

It's very unlikely that radials would have to be limited to 16A - 20A would almost always be possible, and maybe even 25A if you could find appropriate 25A RCBOs for your CU. Having a 20A ring is silly - 20A radial or 32A ring.

The CU will be mounted on Fire Rated chip board with sufficient holes for the cables, the FR board mounted on 2 battens to create a gap down the back and stand off, and allow rear entry via the knockouts, unsure whether any intumiscent putty should go in the hole.

There is no regulatory requirement for a fire-rated 'back board', let alone intumescent putty, and you can have as many 'unprotected' holes up to 12mm diameter in front, bottom and sides of a CU enclosure, provided only that the remaining material between the holes is 'non-combustible'. Such is the common sense of the regs!

Kind Regards, John

 

[EFLImpudence]

the kitchen rings will be on 20amp.

Rings are only still popular because of the special dispensation of using two 27A (54A) cables on a 32A OPD.
This in itself is virtually pointless today with MCBs/RCBOs but the regulation still caters for 30A rewireable fuses - which must have larger than necessary conductors because of their slow fusing factor.

You would not be allowed to do the equivalent with a 20A OPD, so totally pointless.

 

[ericmark]

The ring final (Protected to 32A design current 26A) has the possibility of using 106 meters of 2.5 mm² cable before the volt drop is exceeded. Where a 20A radial is down to 32 meters of 2.5 mm² due mainly to the way design current is calculated, but it means you need 3 radials to replace one ring final.

I do question the volt drop with so many switch mode power supplies, I think you could exceed to 11.5 volt without any real danger, and although you could draw 20 amp at an end double socket it think it is unlikely except for a kitchen.

But a 20A ring final is only 6A below the design current of a 32A ring final due to way calculated, with 20 amp from furthest socket and rest even distributed, so with 32A for volt drop 20 + 12/2 = 26 amp design current.

Fact that the method of calculation is not in BS7671 does not help, only way I found out was asking at IET meetings when I could not replicate the calculations for 106 meters even using correction factors.

 

[bernardgreen]

I think you could exceed to 11.5 volt without any real danger, and although you could draw 20 amp at an end double socket

11.5 V and 20 A equates to 230 Watts heating up the cable, no real danger of the cable getting hot as the heating is spread over 106 metres

Bur wastage of energy

The power from the consumer unit is 230 x 20 = 4,600 Watt

useful power 4,370 W (95%)
wastage 230 W (5%)

 

[JohnW2]

11.5 V and 20 A equates to 230 Watts heating up the cable, no real danger of the cable getting hot as the heating is spread over 106 metres

Indeed. If things are done as they should be, there should never be a problem of the 'cable getting too hot' due to heat being dissipated in it, since the 'maximum current-carrying-capacities' we work with (for particular 'installation methods) when selecting minimum cable sizes are designed to limit the 'energy dissipated per metre of cable' (hence, approximately, cable temperature) to a level which is safe for the cable.

... Bur wastage of energy. The power from the consumer unit is 230 x 20 = 4,600 Watt. ... useful power 4,370 W (95%) ... wastage 230 W (5%)

True but, as above, even if the 'wastage' approached 100%, a correctly-sized cable would still not be at risk.

As with incandescent lighting, at least some of the 'wasted' energy will heat the fabric and/or air of the building, so will not all be actually 'wasted' during the Winter. One also has to consider the fact that if one wanted to markedly reduce the amount of energy wasted by heating cables (which must tot up to a very large amount over a whole country) one would have to increase usage of copper roughly proportionately to that decrease (again, an enormous amount of additional copper {and the energy/environmental pollution etc. involved in obtaining it}).

One should also remember that very few circuits are loaded (ever, or for lengthy periods) to anywhere near their theoretical maximum (the current using for 'calculating VD'), so the 'energy wastage' is far less than calculations like yours might suggest.

In terms of function of connected equipment, and despite what eric often says about traditional (inductive 'ballast') fluorescent lighting, I do think that it would be extremely rare for VD in cables to have any appreciable effect on function of anything. In any event, although I realise that one has to consider possible 'worst case scenarios, many/most of us seem to normally have supply voltages which are more than 10% above the 'permitted minimum (216.2V) - so, for us, the 'recommended' 5% maximum VD is unnecessarily restrictive.

Kind Regards, John

 

[ericmark]

So volt drop limits are not so old radios don't produce a mains hum or motoris don't stall it is just to limit energy loose due to cables producing heat!