Cable Grouping factors and Table 4C1

[eveares]

So following having a read the other day of both the on site guide and the full BS7671 regs that I have, it dawned on me how much of a impact grouping of cables as per table 4C1 has, and how I suspect that what my electrician did back in 2013, violates the regs.

I'm trying to work out that if doing things by the big book per se, (1) if my own calculations are correct and my electrician back in 2013 violated the regs, (2) If I have made a mistake in my calculations, or (3) if there is some exception for short runs of cable, say 2 meters or so where they run up from the DB into the loft before they get disubstituted around the house, not that I can find any such exceptions in the regs.


So see image below for my exact calcs, but having 10 circuits in my house and with all circuits going through around 2 meters of 50x50mm trunking up from the DB into the loft, I can't help but feel that the regs may have been violated if doing things to the letter? Where am I going wrong, or am I??

Based on my own calcs and accounting for the omission of the two circuits that have a load <30% of their tabulated grouped rating, I work out that due the grouping of 8 cables that the my two 32A circuits should have had cable rated for 61Amps.....That's 10mm² cable if clipped direct all the way using method C (i.e. not in the 2 meters of trunking) and not accounting for other factors.

And yes, I am also aware that trunking (Method B) de-rates the 4mm² T&E feeding my kitchen sockets to 30A which is less than the 32A of the MCB.


Help me, I must surely be making a silly mistake somewhere with the grouping factor!?

Excel spreadsheet above is linked here: https://www.grottotree.com/public/Electrical-Diversity-Calcs-To-Share.xlsx


Regards: Elliott.

 

[securespark]

Nothing to do with the calcs, but why are the lighting circuits 1.5?

 

[EFLImpudence]

Where am I going wrong, or am I??

Well - for a start you are not considering time which is a factor in the calculations and the extremely conservative CCC ratings stated in BS7671 which means that virtually no consideration is given to grouping in domestic installations.

Do the cables in the trunking ever get too hot to touch?

Based on my own calcs and accounting for the omission of the two circuits that have a load <30% of their tabulated grouped rating, I work out that due the grouping of 8 cables that the my two 32A circuits should have had cable rated for 61Amps.....That's 10mm² cable if clipped direct all the way using method C (i.e. not in the 2 meters of trunking) and not accounting for other factors.

The cables must be badly damaged, then.

And yes, I am also aware that trunking (Method B) de-rates the 4mm² T&E feeding my kitchen sockets to 30A which is less than the 32A of the MCB.

Conduit does not mean derating if you don't use it.

Help me, I must surely be making a silly mistake somewhere with the grouping factor!?

Perhaps the OSG is giving you wrong advice; at least it is good at that. Throw it away.

I do not really understand your calculations.
For example the hob with 16.6A (after diversity) on 6mm² cable (47A CCC) is surely almost the 30% necessary for discounting let alone counting as 71A?
You have stated the current after diversity on circuits but have not taken it into account for the grouping calculations.

The lighting circuits are both on 20A cable with 1.9A and 1A - that's not going to get warm - yet you only discounted one of them.


On a mathematical point, it does not affect the subject but, why are your calculations not quite right?
E.g. 30% of 6 is 1.8; not 1.755 and 30% of 32 is 9.6; not 9.585.

 

[eveares]

Nothing to do with the calcs, but why are the lighting circuits 1.5?

Because that's what my electrician who did the 1st and 2nd fixing used back in 2013. It was a very long time ago now, but the main/1st electrician never came back to finish it all off due to financial reasons due to getting stitched up by my builder, so my builder had to use someone else to sign it all off...naughty I know. (The contract was with the main builder, not the electrician.)

 

[eveares]

Do the cables in the trunking ever get too hot to touch?

Not that I've ever noticed, but BS7671 bases things on a worse case scenario from what I know.

 

[EFLImpudence]

Not that I've ever noticed,

Well, there you go.

but BS7671 bases things on a worse case scenario from what I know.

No, that's what the OSG does.

BS7671 is for designing installations yourself.

 

[eveares]

The cables must be badly damaged, then

Why damaged?? I was just saying that the tabulated load for 8 cables grouped together where two of them are 32A circuits, would require cables capable of supporting 61Amps for those 32A circuits based on the formula given in the regs.

i.e. For the 32A kitchen sockets, 0.3 (30%) x 0.52 (adjusted correction factor for 8 cables) x 71 (Previous Tabulated load) = 61A cable needed

Conduit does not mean derating if you don't use it.

Except all the cables do use it, if only for around 2 meters.

Perhaps the OSG is giving you wrong advice; at least it is good at that. Throw it away.

I'm just trying to calculate and do the workings by the book per se, not wanting to factor in the possibility that the regs are highly conservative or pessimistic.

The lighting circuits are both on 20A cable with 1.9A and 1A - that's not going to get warm - yet you only discounted one of them.

The load may be far less than the cables rating, but take the downstairs lights for example; the expected load for that circuit in 2013 (before I had LED's) after Diversity was 3.47A, and it's tabulated rating was 13.3A for a 6A circuit (see below).

6 (MCB rating) / 1 (30c Temp) *0.45 (10 cables grouping factor) * 1 (thermal insulation factor) * 1 (not a BS 3036 fuse) = 13.33 Tabulated load

Working out 30% based on the formula 0.3xCgCt assuming the initial group of 10 cables / 0.3x0.52x13.3 = 2.07A for 30% of it's Tabulated rating.

This 30% (aka 2.07A) is < than the expected load of 3.47A after diversity has been applied, and as hence this circuit can not be excluded from the group rating as it's load of 3.47A after Diversity is greater than 2.07A which is 30% of the circuits tabulated rating of 13.3A.

Hope that makes sense.

 

[eveares]

No, that's what the OSG does.

BS7671 is for designing installations yourself.

Maybe, but I have just compared table 4C1 and the notes under it in the on site guide and in the full BS7671 regs book, and they are exactly the same including the notes under the table, with the exception where as the full reg book provides correction values for groups of cables of 16 and 20 cables, where as the on site guide does not.

 

[EFLImpudence]

Why damaged?? I was just saying that the tabulated load for 8 cables grouped together where two of them are 32A circuits, would require cables capable of supporting 61Amps for those 32A circuits based on the formula given in the regs.

Yes, but the reason for derating the cable is to prevent damage.
As it was not done then presumably the cables would get damaged.
They have not therefore the derating must be unnecessary.

i.e. For the 32A kitchen sockets, 0.3 (30%) x 0.52 (adjusted correction factor for 8 cables) x 71 (Previous Tabulated load) = 61A cable needed

But that is obviously complete rubbish - because the cables have not been damaged.

Except all the cables do use it, if only for around 2 meters.

Then that is the only area where this derating might be necessary and as you have said, it is not.

I'm just trying to calculate and do the workings by the book per se, not wanting to factor in the possibility that the regs are highly conservative or pessimistic.

Then you must be applying false criteria.

The load may be far less than the cables rating, but take the downstairs lights for example; the expected load for that circuit in 2013 (before I had LED's) after Diversity was 3.47A, and it's tabulated rating was 13.3A for a 6A circuit (see below).

The cable has a CCC of 20A.

6 (MCB rating) / 1 (30c Temp) *0.45 (10 cables grouping factor) * 1 (thermal insulation factor) * 1 (not a BS 3036 fuse) = 13.33 Tabulated load

Working out 30% based on the formula 0.3xCgCt assuming the initial group of 10 cables / 0.3x0.52x13.3 = 2.07A for 30% of it's Tabulated rating.

This 30% (aka 2.07A) is < than the expected load of 3.47A after diversity has been applied, and as hence this circuit can not be excluded from the group rating as it's load of 3.47A after Diversity is greater than 2.07A which is 30% of the circuits tabulated rating of 13.3A.

You do not need to count all ten cables.
The only one that needs consideration is the kitchen radial, so does even that get warm?

Hope that makes sense.

Depends what you mean by sense.

 

[JohnW2]

Well - for a start you are not considering time which is a factor in the calculations and the extremely conservative CCC ratings stated in BS7671 which means that virtually no consideration is given to grouping in domestic installations.

Quite so. I've never really been able to make much sense of the 'grouping factors in relation to domestic installations - and I suspect I'm far from alone since, as you say I strongly suspect that almost no-one even considers them in domestic installations.

As you imply, the crucial issue is 'diversity'. Common sense suggests that if 'grouping' is ever going to be an issue, it would be when multiple 'grouped' cables carry substantial currents (a substantial proportion of their CCC) simultaneously and for substantial periods of time - and that is just not going to happen hardly ever in a domestic setting, in which several; of the entire circuits will carry zero or near-zero currents for a lot of the time. Things may well be different in non-domestic situations, particularly industrial, but that's a different matter.

There is, as was implied in the OP, also the issue of the .lengths of cables subject to grouping - which, in a domestic setting, is most like to be seen for a very short distance close to the CU. At worst, it's only that little bit of cable that needs 'de-rating' but, in practice, I suspect that short lengths of grouping also reduce the 'effects' of grouping (e.g. by heat conduction along the conductors).

When one thinks about it, I think the only way in which a cable can be influenced by others close to it would be by those other cables, if they were carrying high currents, raising the ambient temp as seen by the cable one was considering - but, again, that needs high current for long periods in the 'other cable(s)'.

 

[flameport]

So see image below for my exact calcs,

Some conclusions from that:

That spreadsheet doesn't contain any calculated values for the circuits other than column E, and that is based solely on whatever 'Initial Tabulated Grouped Rating in Amps' is supposed to be. The rest are just numbers that have been entered from somewhere or other.

The 'required tabulated current' section contains factors which are 1 - they can all be ignored. The calculation is just It / Cg, or for the 32 amps and 0.52 factor it's just 32/0.52 = 61.5

There are various calculations here which are being jumbled together to produce nonsense.

i.e. For the 32A kitchen sockets, 0.3 (30%) x 0.52 (adjusted correction factor for 8 cables) x 71 (Previous Tabulated load) = 61A cable needed

The 30% rule is to determine whether a cable should be included in the grouping calculations. It is NOT the size of cables required.
It's from Note 9 of Table 4C1.

For the 4mm² circuit, the CCC is 37 assuming flat cable method C, and if using that 0.52 factor:
0.3 x 0.52 x 37 = 5.77
If the load current is more than 5.77 amps, cable is included. If less, then it can be ignored and the calculation for the remaining cables uses a correction factor for fewer cables, as if this one did not exist..

6 (MCB rating) / 1 (30c Temp) *0.45 (10 cables grouping factor) * 1 (thermal insulation factor) * 1 (not a BS 3036 fuse) = 13.33 Tabulated load

This one is for the size of cable required, and is what you have got in the upper part of the spreadsheet - It / Cg.
For this 6 amp circuit the answer is indeed 13.33, which is just 6 / 0.45. The other factors of 1 are irrelevant.
What this means is that the cable must be rated to 13.3 amps or more.
As 1.5mm² is rated to 20 amps (4D5, method C), and 20 is more than 13.3, that cable is acceptable.

If you do this for the other circuits you will find that they are not acceptable and instead, massive jumbo sized cables will be required.

However this entire thing falls to pieces as all of this is based on circuits which are loaded up to 100% continuously. See the note below 2.3.1 on page 422, and also general note 1 regarding current ratings on page 432. (both included below for those who don't have the book available)
None of those domestic circuits will be, so all of these calculations are entirely irrelevant.

All of this is based on heating of cables, and where cables are lightly loaded and/or loaded intermittently, the heating effect is far less than for cables which are fully loaded for long periods.

The only likely domestic examples which would be valid for these would be EV charging, where a 32A circuit is likely to be loaded up to 32A for hours at a time, or things such as a hot tub heating from cold, or a hot water cylinder with an immersion heater on for a few hours at a time. For those, the calculations could be used but in reality they won't be as the answer is to install those cables spaced apart rather than all bundled together.

Even for commercial or industrial circuits, the Cg factors can be useless slop. Such as a passenger lift which when moving will draw full load current for a time - but rather obviously it will be stopped for significant amounts of time with no current at all, and just as with the domestic circuits, blindly applying the factors suggested will result in cable sizes which are grossly oversized.

 

[eveares]

That spreadsheet doesn't contain any calculated values for the circuits other than column E

The values is column D were created using the calculator at the top and then manfully entered into the cells D14 to D23. The values in column H were calculated in the same way, but using the corrected grouping factor accounting for less cables.

The 30% rule is to determine whether a cable should be included in the grouping calculations. It is NOT the size of cables required.
It's from Note 9 of Table 4C1.

Well I did exclude two circuits for being <30% of their tabulated load, and then re-calculated the remaining circuits using the corrected and higher numbered correction factor; but are you saying that those remaining circuits that have a load > 30% of their tabulated load, do not need cables sized according to the result of their amended tabulated load?

If so, what point is it in producing a corrected tabulated load which results in a high current rating?

If you do this for the other circuits you will find that they are not acceptable and instead, massive jumbo sized cables will be required.

That's exactly what I did originally as far as I am aware. But of course, my cables are not jumbo sized.

However this entire thing falls to pieces as all of this is based on circuits which are loaded up to 100% continuously. See the note below 2.3.1 on page 422, and also general note 1 regarding current ratings on page 432. (both included below for those who don't have the book available)
None of those domestic circuits will be, so all of these calculations are entirely irrelevant.

Yea I followed the principle in this video (

) and following the same principle, came to stupid results in that I would need stupid sized cables if doing thigs this way.

However this entire thing falls to pieces as all of this is based on circuits which are loaded up to 100% continuously. See the note below 2.3.1 on page 422, and also general note 1 regarding current ratings on page 432. (both included below for those who don't have the book available)
None of those domestic circuits will be, so all of these calculations are entirely irrelevant.

So in hindsight, one can often ignore the grouping factors in domestic settings where EV's are not involved. I also see the onsite guide under table 4C1 says (1) about the cables being equally loaded, which in most domestic situations, they won't be.

I just hope that a council or NICEIC inspector would agree with the reasoning of cables not all being loaded to 100% all of the time (i.e. Notes under 2.3.1) and would not instead make you do things by the OSG and use stupid sized cables to sign off a hypothetical rewire/new build for example.

 

[JohnW2]

I just hope that a council or NICEIC inspector would agree with the reasoning of cables not all being loaded to 100% all of the time (i.e. Notes under 2.3.1) and would not instead make you do things by the OSG and use stupid sized cables to sign off a hypothetical rewire/new build for example.

I think your hope/wish is already with us. As has been said, I don't think hardly anyone even things about grouping factors in domestic installations, let alone does any calculations about them. Hence, if your implied fears were justified, there would probably be very few domestic installations which would 'pass' in the eyes of "a council or NICEIC inspector"!