Electric Cable Routing - Grouping Woes

Discussion in 'Electricians' Talk' started by derek_bartram, Nov 21, 2023.

  1. derek_bartram

    derek_bartram Member

    Good morning everyone,

    Looking for some help on how to route cables for my house renovation (I'm at planning phase, and will be getting Building Control to sign off plans/inspect)...

    The consumer unit is in a difficult position to work with, which means two large bundles of cable need to cross rooms - I can't use the usual 150mm from ceiling due to obstacles, the roof is not practical to access, and metal earth trunking through the walls is problematic.

    Naively, I assumed some metal earthed trunking chased into the floor would solve the problem (Reference Method B?)

    One of the bundles, for example, was to include:
    • 2 * 2.5 TC&E for ring 1
    • 1 * 2.5 TC*E for ring 2 (return comes elsewhere)
    • 1 * 10 TC&E for industrial 32A socket (think garage)
    • 1 * 10 TC&E for solar
    • 1 * 1.5 4-core for smoke detectors
    • 1 * 2.5 TC&E for main lighting feed (i.e. CU to switches)
    • 1 * 1.5 TC&E switched lighting circuit (i.e. switches to bulbs)
    My cable sizes were based on 'the standard sizes'. Where I'm getting into trouble is when calculating the cable sizes with respect to grouping in the trunking...

    The correction factor (Cg) for 2 grouped cables is 0.8, making 2.5 TC&E unsuitable if my 50x50 trunking had only the first two circuits above in it...

    How do practically & economically get the above cables from one side of the room to the other, without digging a foot wide trench and buying a ton of trunking?
     
  2. Coloumb

    Coloumb Screwfix Select

    Grouping is rarely a factor in domestics as the cables don't usually carry their maximum rated load.

    I mean this in a non-critical way but the way you have sized some of these ctt's doesn't really make sense, ie 10mm for a 32a socket and 2.5mm for lighting, which would suggest to me you lack the necessary skills to be able to complete this install yourself. Have you talked to BC about this re-wire as they would be unlikely to certify it without some convincing that you know what your doing.
     
  3. derek_bartram

    derek_bartram Member

    Could you elaborate on what the criteria for 'rarely' is please? I'm not disputing you, simply learning and trying to correlate your advice with the BS7671 documents - from my reading and understanding (which could well be wrong), but there doesn't appear to be any wiggle room; if you put the cables in trunking under the floor, then the grouping correction factor applies.

    The circuit sizing, is within the bounds of normal sizing:
    The lighting circuits, from switch to bulb are all the more common 1.5, however I've opted for 2.5 from CU to switches down to a) expected circuit length factors (i.e. voltage drop), and b) some expectation of grouping issues (which ironically does make the original question easier, as the calculations to my understanding need it).
    The garage connection, using Method 102 (as it would further down the circuit), and applying Table 4D5, gives the minimum conductor size for TC&E as 10mm^2... which I've specified.​
    What about that makes you doubt my skills?
     
  4. WH55

    WH55 Screwfix Select

    I assume his point is, in part, that you must have a heck of a big house to need 2.5mm T&E for a lighting circuit to cope with volt drop.
     
  5. derek_bartram

    derek_bartram Member

    Ok, fair enough, voltage drop alone isn't sufficient to justify a bump to 2.5mm T&E.... however, put it in a trunk with 4 other circuits, and Cg becomes 0.6, and suddenly it *appears* that 2.5mm is required.

    Either grouping factor doesn't apply, as Coloumb suggests, but why?
    Or grouping factor does apply, in which case the original question stands?

    I'm looking for an answer based on BS7671 and the question posed, not a critique of why I have the circuits I have.... please :)
     
  6. malbec4

    malbec4 Well-Known Member

    As pointed out by @Coloumb, grouping factors in BS 7671 only apply when the cables are carrying the maximum rated load. If this happened in real life, trunking and conduit could hardly be used.

    If you ran all the cables that you propose to their maximum, it works out at about 218 amps. If that was the case, you would have more to worry about than grouping factors.

    If you really want to explore the wiring regs, account has to be made for space factors in trunking. Reg no 522-08 makes reference to it, but does not give much detail. Guidance Note 1 has the details and should be consulted. In trunking 45% of the area within the trunking should be free space. This would be of greater relevance than grouping factors.
     
  7. derek_bartram

    derek_bartram Member

    Am I being stupid here, because I can't see anything to do with 'only apply when at maximum load'.

    From BS7671: Current-Carrying Capacities and Voltage Drop for Copper Conductors:

    I'm paraphrasing here, but this is what it states...

    lb = design current = 32A for socket rings
    ln = overcurrent device rating, where ln >= lb = 32A

    lt = tabulated current-carry capacity of cable
    Cg = cable grouping factor (Table 6C / Method 3) = 0.8 for 3 circuits

    lt >= ln / (Ca * Ci * Cg * Cr)
    lt >= 32 / (1 * 1 * 0.7 * 1)
    lt >= 45.7

    Table 6E1, Reference method 4, 1 two-core cable / single phase:
    2.5 section = 18.5A
    4 section = 25A
    Nowhere does it state to use expected load instead of design current?

    "If you ran all the cables that you propose to their maximum, it works out at about 218 amps."
    Cable selection is different from circuit diversity - this is why the theoretical capacity can exceed to total fuse allowance. An incidentally, I get 93A when you apply the rules for diversity in Appendix 1: Maximum Demand & Diversity
    "space factors in trunking"
    Space factors in trunking has very little to do with cable selection, and Appendix 5: Cable Capacities of Conduit and Trunking states quite specifically:
    "Only the ease of drawing-in is taken into account. The electrical effects of grouping are not. [...] Cable sizes have to be increased"

    Also, (and I specifically note for interest, more than an intent to discredit you):
    "In 1996, IEE Guidance Note 1 provided further background to the tables issued in the Fourteenth Edition of the IEE Wiring Regulations and acknowledged that the previous space factor of 40% for conduit and 45% for trunking was an arbitrary value, later was shown to be inappropriate."​
     
  8. malbec4

    malbec4 Well-Known Member

    It refers to tabulated cable sizes, like singles. In cases where the cable size is not listed, twin & earth for instance. the 45% rule still applies.
     
  9. malbec4

    malbec4 Well-Known Member

    Oops. I added it up wrong. Must remember to use a calculator next. Applying a bit of diversity would help as well.

    I'm not sure what expected load means.

    I feel a Tuesday beer coming on. I'll get back to you. You may have some good points, but I can't be bothered looking them up. I will tomorrow though.
     
  10. The Happy Builder

    The Happy Builder Screwfix Select

    You are trying to work at a level that most electricians are not actually qualified to do, you should really do the City and Guilds Level 4 Design, Erection and Verification course to become qualified or use standard circuits from the IET On-Site Guide.

    Domestic electricians are generally expected to have and use a copy of the IET for circuit design, I suggest you buy and refer to the IET OSG and don’t try to ram too many cables into some trunking.

    Some years ago one of the electrical text books written by Brian Scaddan gave an example of a circuit heater design calculation that demonstrated that due to bad design of the installation a 16.0 mm twin and earth cable was required to supply a 3.0 kW immersion heater rather than a 1.5 mm which would be adequate if the installation was designed appropriately, because the cable was badly routed through insulation.

    The true purpose of that example in the text book was to instil the importance of ensuring cable routes are selected correctly along with installation methods.

    In another post you mention galvanised steel back boxes not having knockouts for 25 mm conduit, that’s because no manufacturer would expect you to be using 25 mm conduit, likewise you will find you cannot generally wire lighting circuits in 2.5 mm twin and earth, because the terminals are made for 1.0 and 1.5 mm.

    You need to completely rethink what you intend to do and follow general custom and practice.
     
  11. arrow

    arrow Screwfix Select

  12. The Happy Builder

    The Happy Builder Screwfix Select

    Obviously that is not a 25 mm back box, because it is physically impossible to have a 25 mm knockouts in the side of a 25 mm box ;)

    A 47 mm back box is actually intended for a shower or cooker circuit fittings, which will be wired in 6.0 or 10.0 mm, not sockets and lighting which will be wired in smaller cables and will not require 25 mm conduit, which seems to be the original posters intention, also conduit and trunking is intended for singles not twin and earth.

    It appears that Screwfix 25 mm and 35 mm galvanised back boxes only have 20 mm knockouts, which is not really surprising.
     
    Last edited: Nov 21, 2023
  13. derek_bartram

    derek_bartram Member

    Thank you all for your continued input on this, I genuinely appreciate it!

    Please read the below as me trying to understand your answers and explain why I don't answer them.

    "I added it up wrong."
    You added it up, I can only assume correctly, however your point was 'it's far too much'. I applied diversity to get 93A. ​

    "I'm not sure what expected load means."
    You said "grouping factors in BS 7671 only apply when the cables are carrying the maximum rated load", which implied I should be considering 'expected load' and if it were less than 'maximum rated load'.

    Ps. However comes up with a good solution is more than welcome to meet me in Penrith for a Tuesday beer or two on me!
    Pps. And thank you, I would be really grateful if you could - I'm clearly out of my depth on this!​

    "buy and refer to the IET OSG and don’t try to ram too many cables into some trunking"
    I have it and I'm using it - 2.5mm lighting wiring not withstanding, that's how most of the above comes up.
    The difficulty I have is: my house is very old and it's not possible to route in the usual way (e.g. through partition walls, top of walls, etc.) - hence, I either need to be able to plan and calculate an alternative. I simply cannot follow the standard case, because the building doesn't allow it, I absolutely would otherwise.
    "A 47 mm back box is actually intended for a shower or cooker circuit fittings"
    Completely understand that, however in trying to solve the floor problem it creates new problems.
    <and as ever, correct me if you see something wrong here, I'm not 100% sure on any of this>
    Going across a floor, within 50mm of the surface, means earthed metal conduit/trunking. More than 2 cables in trunking screws the grouping factor, which means either a) larger cables, or b) more routes. Larger cables is expensive and, as you quite rightly point out, makes it difficult to get fittings which can accept the larger size. More routes = more trunking/conduit = more expensive, bigger chases, and harder to route. Flexible metal conduit would probably be easier to route than rigid, however would need the end terminated - noting the usual size is 20 or 25mm.
    I could use a larger back box and terminate either size, however 48mm is a convenient size in my case: my outside walls will include insulation, then 38mm battens, and then 12mm plasterboard. With the right mounting, I can get a 48mm backbox rigid to the front of the plasterboard.
    Please do keep the comments coming! Thanks.

     
  14. derek_bartram

    derek_bartram Member

    Thinking about, can I route in 3 conduits...

    Conduit 1: ring 1 & ring 2
    It's three cables, but only two circuits, so Cg = 0.8
    lt >= 32 / 0.8 = 40A
    A 2.5mm cable would provide 21A when directly in insulated wall (worse case on route)
    Conduit 2: industrial socket & solar
    lt >= 40 / 0.8 = 50A
    I would need to upgrade to 16mm cable
    Conduit 3: main lighting feed & switched lighting feed & smoke detectors
    Both lighting feeds are the same circuit, so it's actually only two circuits, so Cg = 0.8
    lt >= 6 / 0.8 = 7.5A
    A 1.5mm cable would provide 15A when directly in insulated wall (worst case on route); strictly I could even use 1mm = 12A
    The question then becomes, can I standardize on 20/25mm conduit?
    From OSG Appendix 5, assuming an fairly extreme 3 bends over 5m...
    1.5mm = Cable factor 22
    2.5mm = Cable factor 30
    16mm = Cable factor 145

    Conduit 1:
    Combined cable factor = 3 * 30 = 90
    In 20mm = maximum allowed factor of 141
    Acceptable with decent margin
    Conduit 2:
    Combined cable factor = 2 * 145 = 290
    In 25mm = maximum allowed factor of 260
    Not acceptable, so would need to use either trunking or 32mm conduit
    Conduit 3:
    Same as conduit 1, only the cable factor is slightly smaller, so would be fine.
    So... I believe I could trunk the two big cables, and use 2 flexible metal conduits for the rest?
     
  15. derek_bartram

    derek_bartram Member

    ***, looks like flexible metal conduits is out and I have to lay solid trunking everywhere :mad:
    "Regulation 522.8.2 which requires a conduit system to be completely erected between access points before any cable is drawn in, refers to conduit systems which are buried in a structure"
    I can't fit solid trunking into the insulated walls, so I'm stuck again. I could help please, I'm really stuck
     
    Last edited: Nov 22, 2023
  16. malbec4

    malbec4 Well-Known Member

    Okay. I have dusted down the page containing the grouping factors for cables in trunking.

    Appendix 4, 2.3.3 & 2.3.3.1

    I knew something odd went on, but couldn't remember. I don't think I've looked at it for many a year. Previously, I was getting mixed up with different installation methods which requires all cable sizes to be the same as well as the load. Which never happens.

    Anyway, it says cables of different sizes and different loading are outside the scope of this appendix. Very helpful.

    But then it comes up with a little formula with a very onerous outcome.

    Cg=1/square root of number of circuits

    In your case this comes to Cg=0.377.

    So, for a 32 amp breaker a cable that can take 84.9 amps is required. So just wire the sockets in 25mm cable!


    A more sensible way of doing it is to make sure the trunking is big enough so that heat can dissipate easily. Apart from the 32 amp socket (load unknown) there doesn't seem to be heavy loads constantly on . Then ignore the grouping factor. It's no different to when a consumer unit is connected up. You can have, say, 36 circuits in a consumer unit and nobody ever derates the cabling.

    You can use twin & earth. Can you not use 6mm for the 32 amp radials? More room in the trunking.

    One last thing. Why 4 core for the smoke alarms? Or did you mean 3 core and earth?

    ps Penrith is a bit too far, even for a beer.
     
  17. derek_bartram

    derek_bartram Member

    Indeed - although given the massive penalty of grouping factor, one assumes you'd never be able to get enough cables in for it to matter!?!

    (obviously a joke, but...) If you do that, you then can't fit the cables in the trunking according to Appendix 5 - damned if you do, damned if you don't

    It's ultimately a massive extension cable to a workshop, but could conceivably peak up there briefly.

    A) How do I work out the required size, and more importantly, B) how do I demonstrate that to Building Control when I submit the plans - the latter point is why I'm so focussed on grouping factors, when your average electrician would dismiss it with engineering judgement (quite reasonably, one assumes)

    I did wonder that, and confess I had forgotten about it... although, in thinking about it more, and having read more regarding grouping near consumer units, that grouping would not apply if wiring was installed correctly. The width of MCB/RCBO would always ensure sufficient separation of a minimum of a cable thickness, and you're meant to space the cables sufficiently when you leave the box - not to say the latter actually happens, but you get the point. I'm not sure Building Control would look kindly on my planning if I used 'everyone else does it' as my design justification :(

    I don't have a physical space problem, I have a grouping problem. Switching the rings to radials also doesn't help, because the calculations are based on circuits and not cables (ok, granted it would help a little, but I have a similar problem elsewhere where it couldn't). Also, you can't get two lots of 6mm into the back of standard cheaper sockets.

    Funny story here... Yes, I meant the latter, however when searching on TLC I couldn't from 3C&E; I emailed them and they told me it's called 4C. 3C&E is a much better name for it, especially when it uses the harmonised colours as standard.

    I'm driving to Cornwall in a few weeks time - I can deliver :)
     
  18. malbec4

    malbec4 Well-Known Member

  19. derek_bartram

    derek_bartram Member

    That's the mistake I made when starting to look into this (same problem for plastic and metal). Those figures are based on BS7671 On-Site Guide, Appendix 5 Cable capacities of conduit and trunking.
    In principal all looks well, except, that section states:
    "Only the ease of drawing-in is taken into account. The electrical effects of grouping are not."
    What your link and the table means, is you can physically put 38 pieces of 2.5 TC&E in the space, however there's nothing to say how much those cables would actually be able to carry.
    If you put 10 cables in 50x50 trunking, then each cable would only be able to support 50% of its normal current capacity (before you even factor for how much current carrying capacity a single cable would be able to take, with respect to installation environment).

    The table is basically meaningless for any circuits which are not massively over-provisioned, in my opinion.
     
  20. malbec4

    malbec4 Well-Known Member

    I realise all that. I was showing you that 50 x 50 trunking could accommodate your cables, not realising that you had already seen that table. But to make sure, go for a size bigger. That's why I suggested 100 x 50. The cables would be lost in that.

    The odd thing about derating for cables in trunking is that trunking sizes are not mentioned. So the derating (as I posted earlier) is the same for 50 x 50 trunking as it is for 300 x 300, with the same amount of cables installed. You need to make sure that the trunking is big enough. As I said, there is plenty of room, just ignore derating for trunking.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice