Suitability of SWA armouring as CPC

IAmSparkytus! · Oct 15, 2022

Use SWA as CPC, a guide to the acceptability of steel wire armour use as CPC.

SWA as the CPC? In most case it is fine to use the SWA as a CPC.

gadsolutions.biz

Anyone know how reliable the tables on this website are for determining whether the armouring of a cable can be used as it's earth?

Not too sure how old the page of the website is so could have changed since it was made or could have been wrong to start with I suppose

TIA

westward10 · Oct 15, 2022

My tablet flags that up as a dodgy website.

pc1966 · Oct 15, 2022

I don't get it flagged as dodgy, but maybe my standards are lower...

It reminds me of something I read a while back, probably an IET article or publication. the page says last updated 2012 so not quite current, even though most of the factors remain the same.

IAmSparkytus! · Oct 15, 2022

westward10 said:
My tablet flags that up as a dodgy website.

Definitely not X-rated so you're not missing out I wouldn't worry!

davesparks · Oct 15, 2022

IAmSparkytus! said:
Use SWA as CPC, a guide to the acceptability of steel wire armour use as CPC.

SWA as the CPC? In most case it is fine to use the SWA as a CPC.

gadsolutions.biz

Anyone know how reliable the tables on this website are for determining whether the armouring of a cable can be used as it's earth?

Not too sure how old the page of the website is so could have changed since it was made or could have been wrong to start with I suppose

TIA

That table has been knocking around in various forms for years and is based on the simplified calculation of CPC size. As a result it will be correct where it says that the armour is a suitable size, but if you calculate using the adiabatic equation you will find that where the table says the armour is unsuitable it is in actual fact almost always suitable.

This all assumes the armour is only being used as a CPC and is not required to also be a main bond, in which case it will almost certainly be unsuitable.

pc1966 · Oct 15, 2022

Nope, checked again and for 90C thermosetting k1 is 100 from Table 43.1 and k2 is 46 from Table 54.4 so here are the values I get, assuming those k values and assuming Fe is 8 times poorer than Cu for conductivity:

Live Cu	Min CPC Cu Table 54.7	SWA adiabatic eq 54.7	CPC PME Table 54.8	PME SWA
1.5	1.5	3.3	10	80
2.5	2.5	5.4	10	80
4	4	8.7	10	80
6	6	13.0	10	80
10	10	21.7	10	80
16	16	34.8	10	80
25	16	34.8	10	80
35	16	34.8	10	80
50	25	54.3	16	128
70	35	76.1	25	200
95	47.5	103.3	25	200
120	60	130.4	35	280
150	75	163.0	35	280
185	92.5	201.1	50	400

pc1966 · Oct 15, 2022

What you see is the PME main bonding cases is far more demanding, basically as it might be carrying tens or hundreds of amps for hours under open-PEN fault conditions and is not met by any of the SWA cable construction. So it is really a case of external CPC in parallel, or use an extra core in the SWA if that is the means by which any bonding is returned to the DNO connection point.

EDIT: Also note the extra note in 544.1.1 with AM2 about cables to other building not having to be fully to this in some cases.

pc1966 · Oct 15, 2022

deleted deleted deleted

cliffed · Oct 19, 2022

pc1966 said:
Nope, checked again and for 90C thermosetting k1 is 100 from Table 43.1 and k2 is 46 from Table 54.4 so here are the values I get, assuming those k values and assuming Fe is 8 times poorer than Cu for conductivity:

Live Cu
Min CPC Cu Table 54.7
SWA adiabatic eq 54.7
CPC PME Table 54.8
PME SWA
1.5
1.5
3.3
10
80
2.5
2.5
5.4
10
80
4
4
8.7
10
80
6
6
13.0
10
80
10
10
21.7
10
80
16
16
34.8
10
80
25
16
34.8
10
80
35
16
34.8
10
80
50
25
54.3
16
128
70
35
76.1
25
200
95
47.5
103.3
25
200
120
60
130.4
35
280
150
75
163.0
35
280
185
92.5
201.1
50
400

The end list… is it for the swa used for that circuit..16mm being 80 mm & confirms that the armouring is compliant

pc1966 · Oct 19, 2022

cliffed said:
The end list… is it for the swa used for that circuit..16mm being 80 mm & confirms that the armouring is compliant

No, it means if the SWA armour is your main bond for PME situations it needs to be 80mm section to match 10mm copper as specified in Table 54.8. That is only met for big cables such as (I think) 35mm 4C, 50mm 3C, and 70mm 2C or larger. That 10mm PME requirement is why often folks use 10mm 3C for an out-building with extraneous parts as then the copper CPC meets it.

The PME size I used is based on the incoming cable with Table 54.8, for a sub-main you might well have much larger conductors if the length is significant just to meet VD.

But under AM2 there is some relaxing of that if its a sub-main to other buildings in a group but would need to read the details of it.

cliffed · Oct 19, 2022

pc1966 said:
No, it means if the SWA armour is your main bond for PME situations it needs to be 80mm section to match 10mm copper as specified in Table 54.8. That is only met for big cables such as (I think) 35mm 4C, 50mm 3C, and 70mm 2C or larger. That 10mm PME requirement is why often folks use 10mm 3C for an out-building with extraneous parts as then the copper CPC meets it.

The PME size I used is based on the incoming cable with Table 54.8, for a sub-main you might well have much larger conductors if the length is significant just to meet VD.

But under AM2 there is some relaxing of that if its a sub-main to other buildings in a group but would need to read the details of it.

Thank you understand now the poor conductivity of the armouring.

loz2754 · Oct 19, 2022

pc1966 said:
But under AM2 there is some relaxing of that if its a sub-main to other buildings in a group but would need to read the details of it.

There is a whole thread about this, also started by @cliffed , the conclusion of which was that this amendment applies only where PME conditions do not apply.

Power point in external glasshouse

Client wants 2g socket in greenhouse, obviously IP rated,will be supplied from Rcd, the greenhouse consists of metal frame work on the ground. Does this require MPBC

www.electriciansforums.net

davesparks · Oct 19, 2022

pc1966 said:
Nope, checked again and for 90C thermosetting k1 is 100 from Table 43.1 and k2 is 46 from Table 54.4 so here are the values I get, assuming those k values and assuming Fe is 8 times poorer than Cu for conductivity:

Live Cu
Min CPC Cu Table 54.7
SWA adiabatic eq 54.7
CPC PME Table 54.8
PME SWA
1.5
1.5
3.3
10
80
2.5
2.5
5.4
10
80
4
4
8.7
10
80
6
6
13.0
10
80
10
10
21.7
10
80
16
16
34.8
10
80
25
16
34.8
10
80
35
16
34.8
10
80
50
25
54.3
16
128
70
35
76.1
25
200
95
47.5
103.3
25
200
120
60
130.4
35
280
150
75
163.0
35
280
185
92.5
201.1
50
400

In your column headings I think that the third column should be selection in accordance with table 54.7 and not adiabatic equation shouldn't it?

pc1966 · Oct 19, 2022

davesparks said:
In your column headings I think that the third column should be selection in accordance with table 54.7 and not adiabatic equation shouldn't it?

It is badly explained in my post but by "adiabatic eq" I meant the adiabatic equivalent conductor to the copper one.

So from Table 54.7 it has you list of CPC=line to CPC=line/2 sizes, and then the suggestion that for fault handling equivalent using other types of material you scale the size by the ratio of the two adiabatic k-factors (k1 & k2). This leads to a conductor that has the same short-term overload handling as it takes in to account both the higher resistance of other conductors like steel, but also that as you make the conductor bigger to reduce that resistance penalty, it also increases the heat-absorbing capacity.

If the only difference was material resistance, and not any other thermal factor, then you would need the square root of the conductivity bigger. For example, if you had a poor copper version that was identical to copper in every way but it had 4 times the electrical resistance, but time you make it sqrt(4) = 2 times larger to drop the resistance to only double the normal copper resistance, you also have double the metal to adsorb the energy and so you reach the same post-fault temperature form given I2t. If it was 9 times less conductive then by time you make the CSA 3 times bigger you have 3 times the I^2R heating but also 3 times the mass, etc, etc.

That is far less of a problem than the open-PEN fault case where the CPC has to sustain a given high current to whatever is bonded (say metal service pipes back to the substation via another TN-C-S bond) as then your temperature rise is given only by I^2R and the thermal environment of the cable. So to make steel armour run at the same ultimate temperature as a copper CPC you need its R to be the same (more or less, a larger cable will generally dissipate heat better as more surface area) so instead of the k1 & k2 values being used you are looking at the material resistivity, which is about 8 times instead of the k-factors of 100/46 = 2.17