IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Ring Circuit Grouping factor Cg
Topic Summary: 2396
Created On: 18 April 2017 07:45 PM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 18 April 2017 07:45 PM
User is offline View Users Profile Print this message



jamesy81

Posts: 5
Joined: 07 April 2017

Just started this course, also been studying the design guide for a couple of weeks now..

I am feeling a bit bewildered when it comes to grouping factors, relating to ring circuits, using single cores within trunking.

According to BS 7671..

Iz /> 20A (Reg 433.1.204)
Iz /> In (Reg 433.1.1);

In </ Iz
In </ 20
Assuming In = 20A

4mm single cores are practical and assuming all other correction factors = 1..

It = 32 (Table 4D1A Ref B)
It /> In/Cg (Appendix 4; 5.1.1);

Cg /> In/It
Cg /> 20/32
Cg /> 0.625

Or 4 circuits max. That's not a lot of circuits in a trunking.

I keep thinking diversity but as Iz has to be at least 20A I cant see how it can be applied

I think I am completely missing something here, any help would be appreciated
 18 April 2017 09:26 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9107
Joined: 22 July 2004

Actually the effect of trunking on current ratings can be pretty stifling, and of course the rules as written have to assume a worst case.
I don't think you are missing anything other than the fact that in practice, where the loading is known to be light, the grouping thing is quite commonly not applied as per regs.
Quite often rings are not heavily loaded, but having more than one is just a convenience of isolation by room or by zone.
You may care to think you could have had just one ring supplying 1000 square feet of floor area, now if the area is subdivided, but the total load is not changed, so that at any instant, at worst, one ring, and you dont know which one, is fully loaded, but the others are at say 50%.or even less.
Also as Dave Z pointed out in this related thread the regs make no distinction between a large trunking, with lots of airspace, and a smaller one mostly full. Also convection is very different in vertical boxing as opposed to horizontal, but the regs ignore that.

Some real rough installations look like the cables were fitted by shoe-horn and a rugby scrum style push is needed to get the lid on, ideally one should size things to be less than half full, so that by the time there has been some mission creep and your design has been added to by others, it is still usable.

-------------------------
regards Mike
 18 April 2017 09:29 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 15632
Joined: 13 August 2003

I keep thinking diversity but as Iz has to be at least 20A

A conventional ring has a 32A (or 30A) protective device - so if one leg is carrying 20A, the other can't be carrying more than 12A. Or if you prefer, roughly equivalent to both legs carrying 16A.
- Andy.
 18 April 2017 09:46 PM
User is offline View Users Profile Print this message



Legh

Posts: 4008
Joined: 17 December 2004

Iz must not be less than 20A
2.5/1.5mm2 rated at 26A max
doubled in a ring = 52A
therefore use 32A = In
Now go back and look at circuit design from first principles
IL <= Ib <= In <= It <= Iz (433.1.1)

Legh

-------------------------

http://www.leghrichardson.co.uk

de-avatared
 18 April 2017 11:08 PM
User is offline View Users Profile Print this message



broadgage

Posts: 2317
Joined: 07 August 2007

Another factor to consider is that in domestic or similar premises, with a 100 amp maximum single phase service, you cant have more than 3 ring final circuits fully loaded other than briefly.

3 fully loaded 32 amp ring finals is of course 96 amps in total, so with a 100 amp service, ALL other cables in the trunking can be ignored as they cant be carrying more than 4 amps in total.

Similar arguments may apply in larger installations if submains are fused at 100 amps or less.
 18 April 2017 11:47 PM
User is offline View Users Profile Print this message



jamesy81

Posts: 5
Joined: 07 April 2017

Ah I see my mistake - got bit mixed up with all these inequalities,

Of course, In = 32A (or 30A), cant believe I said otherwise, wasn't thinking about what i was writing at all there.

I should have posted this..

Iz /> 20A (Reg 433.1.204)
It = 32 (4mm, Table 4D1A, Ref B)

Referring to,

Iz = It.Cg (Appendix 4)

Therefore,

Cg /> Iz/It
Cg /> 20/32
Cg /> 0.625

Its the same point though,

Using 4mm singles, grouped with more than 4 other circuits causes
the current carrying capacity to fall below below the minimum required 20 A

4 CCTS: Cg = 0.65
5 CCTS: Cg = 0.6

Iz = It.Cg
Iz = 32 x 0.65 = 20.8 (4mm sufficient)

Iz = It.Cg
Iz = 32 x 0.6 = 19.2 (4mm insufficient)

Or the other way if preferred

It /> Iz/Cg
It /> 20/0.65
It /> 30.76 A (4mm minimum)

It /> Iz/Cg
It /> 20/0.6
It /> 33.33 A (6mm minimum)

Thanks mapj1 its mad thinking back to the amount of circuits I've installed like this and never even gave grouping a second thought, totally oblivious.

Apologies again for the nonsense I posted at first guys!

Edited: 18 April 2017 at 11:56 PM by jamesy81
 19 April 2017 01:18 PM
User is offline View Users Profile Print this message



tillie

Posts: 881
Joined: 03 May 2006

Hi , in Electrical installation calculations by BD Jenkins , which I find to be the best book by far for this sort of stuff it has a table for factors to be applied for grouped ring circuits.

The book may be a little dated from 1991 and revised in 2010 but the maths cannot be that different and I would base my calculations on the advice given.

2 ccts = 1.19
3 ccts = 1.03
4 ccts = 0.94
5 ccts = 0.87
6 ccts = 0.82
7 ccts = 0.77
8 ccts = 0.74
9 ccts = 0.71

ie for a system with 4 ring circuits grouped then In/Cg = 32/0.94 = 34a.

Regards
 19 April 2017 02:02 PM
User is offline View Users Profile Print this message



mapj1

Posts: 9107
Joined: 22 July 2004

The conclusion from that table is very similar to the original post- as somewhere between having 4 and 5 fully loaded ring circuits co-routed you have to step up cable size from 4mm to 6mm unless its is some special high temperature cable.
(realising that the missing line at the top of the table would be 1.5 for one circuit 20A cable 32A breaker)
However, in a real system, if you actually had four fully loaded circuits, (so more than 120A total steady loading) you'd probably rather use thinner cable but in oversized trunking or parallel runs of smaller trunking with a bit of an air gap, so as to spread the heat out a bit. This is also the reason for spacing cables out on when loading up baskets or trays.

-------------------------
regards Mike
 20 April 2017 08:06 PM
User is offline View Users Profile Print this message



davezawadi

Posts: 3806
Joined: 26 June 2002

One could look on it a slightly different way, that is that you have a singlre 8mm conductor carrying 32A maximum, and then apply the grouping numbers with half the conductors, which may be a better way with ring circuits, and in fact with a bit of underfilling is more like a 10mm conductor. If the ring is really a ring, don't worry as several circuits will still average out the same way. Now have you got the answer you wanted? Design is a bit like that!

-------------------------
David
BSc CEng MIET
david@ZawadiSoundAndLighting.co.uk
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..