# SWA plus separate cpc

While on a site last week, I noted 25mm2 4-core swa cables being installed on a ladder and for some reason, each one had a 10mm2 separate copper cpc cable tied to the swa. The furthest run is about 50m from a DB where the EFLI is 0.1 ohms.

Just as an academic exercise, I set about trying to establish the current division between the wire armour and the separate 10mm2 copper cpc using the formulae in the IET Design Guide. Factoring in the usual temperature increases, 70 for phase in the swa, 60 for the wire armour and 30 for the separate cpc, the expected impedance was 0.25 ohms with a fault current (using 230) of 1140A dividing to 672A in the wire armour and 826A in the separate cpc.

Data used; R1 at 20C = 0.73mohms, R2a at 20C = 2.1mohms and R2p at 20C  = 1.83mohms. Likely there is no real merit in the pedantic adjustment for temperature, but hey!

If someone felt sufficiently bored and had an old fag packet, would they be kind enough to check my results? Strangely, if one reduces the separate cpc to 2.5mm2 the fault current is 936A dividing to 244A in the wire armour and 727A in the 2.5mm2 separate cpc. That is counter intuitive but perhaps expresses a limitation on the formulae or, much more likely, a limitation on my understanding!

Parents
• A handy guide, but it does not seem to be displaying the tables correctly with the sizes that are a problem in red.

• JP may recognise some of the words on that web page as the same as something he too has circulated on the topic but as you say with the vital formatting to identify the problem sizes slightly stuffed..!!

John does not consider anything more than an ohmic split- which will be OK for smaller SWAs The formulae above are really for when inductance is starting to bite over simple resistance - hence its derivation using test results from 95 and 185mm cables - hence my comment on it really needing some extra notes on the range of applicability.

Mike

• And so it goes on, strapping a supplementary cps to the swa cable, completely dismissive of cost to the client or environmental impact. I asked my 2391 class yesterday, all contractors and all would feel uncomfortable about relying on the wire armour alone. Several confirmed that every consultant-designed job they had ever been on specified the separate cpc. It is now part and parcel of standard practice, like pulling a cpc into a steel conduit wiring system.

However, my concern is more from the academic side. I note that City and Guilds are quite rightly encouraging reference to the IET Design Guide for the 2396 Design Erection and Verification course so as Mike says, some extra notes on the range of applicability would be good rather than a bald set of formulae that come without explanation or example. In fact, had an example been given the errors pointed out by GB would likely have been spotted before publication.

A sample calculation exercise for one of the revised City and Guilds  EV courses comes with their suggested answers. Although it is only swa cable without an external cpc, it uses the following formula from the IET Design Guide;

Now, anyone who has ever sat a City and Guilds exam of this type will know how pedantic the examiner can be on adjusting for operating temperature! The sample answer given raises the R1 and R2 values as normal but ignores the additional reactive component, perhaps because the cable is only 6mm2.

Any way, what I want to establish is (a) is the formula a vector addition or is it just a bit of rough maths ignoring the fact that you ought not to add apples and oranges in the same way we add Rs to Zs in simple loop impedance calculations and (b) the application of a1.1 uplift to the DC resistance of the SWA is to account for the magnetic effect of the conductors within the containment, which the wire armour is not, so if the 1.1 uplift accounts for a change in reactance, perhaps to the phase but applied to the armour, then why further adjust for operating temperature as reactance will not be affected by it.

As a foot note for GB; Being 73 is not an excuse. I remember you well for your sterling contribution to the academic and technical side of our industry over very many years. Please do not duck under the quilt just because of increasing years. Every time I see your name pop up on this forum, I am straight there! You probably do not appreciate your value to light-weights like me!

Oh, and by the way, my neighbour is an electrical contractor and farmer, he is up at 6am every morning to tend to his cattle and off down the road to some other farm to fix milking machines and the like. He has just turned 84!!

• However, my concern is more from the academic side. I note that City and Guilds are quite rightly encouraging reference to the IET Design Guide for the 2396 Design Erection and Verification course so as Mike says, some extra notes on the range of applicability would be good rather than a bald set of formulae that come without explanation or example. In fact, had an example been given the errors pointed out by GB would likely have been spotted before publication.

I think whether an example would have identified the issue depends when the errors were introduced in the process ... basically, equation type expressions are not easy in publishing packages, and there's no ready "export from Word" or similar. As you will have noticed, there are some other issues with the subscripts of one of the formulas here too.

I can confirm that the two formulas in D5-5 (a) and (b) are "swapped", and the formulas (as per PD IEC/TR 50480) should have been provided as follows:

The formulas in the paper Geoff Blackwell refers to eventually founds its way into the UK national annex to PD IEC/TR 50480, which is referenced in the EIDG.

I'm sure the suggestion about an example for the "parallel cpc" will be considered seriously for the next Edition ... but I have some thoughts of my own below that might prompt some discussion.

Now, anyone who has ever sat a City and Guilds exam of this type will know how pedantic the examiner can be on adjusting for operating temperature!

Agreed. Annex M (section M2) of the 5th Edition of the IET Code of Practice for EV Charging Equipment Installations does include an example for armour (alone) used as cpc, and different temperature correction factors are used for the copper and steel ... although the assumption here is that the armour will be at the cable conductor operating temperature before the fault begins, which it is unlikely to be. It also includes the note that reactance is not  adjusted for operating temperature.

And so it goes on, strapping a supplementary cps to the swa cable, completely dismissive of cost to the client or environmental impact.

I have some concerns (apart from getting the maths right) about the validity of the practice of supplementary cpc in all cases:

• We all know that BS 7671 is supposed to cover only single-fault conditions. The debate here, is whether the "single-fault" is a severed cable or not ... because if it is possible to sever the cable and not the external cpc, the external cpc can't be taken into account for that fault. A severed cable (or at least damage to more than one conductor) must be considered as a "single fault"  because we use the three-phase bolted fault current to determine maximum prospective fault current.
• In addition to the above, for direct-buried cables (where unintentional damage of an unseen cable is more likely, and may damage or sever the SWA and not the parallel cpc) Regulation 522.8.10 says the armour has to be suitable for use as a cpc.

But I'm not 100 % decided and I think the designer needs to take the actual risks into account before deciding whether to go ahead with the practice.

Finally, I've heard some say that we can't rely on the SWA as a cpc because of issues with poor termination in glands ... for me, that's a step too far. DO IT PROPERLY (it's not like we don't have all the fancy tools available at very reasonable prices compared to 25 years ago)!

• Graham, I see that you recognise the adjustment of the SWA probably ought not to be at 70C. Fair enough but the correction should be to 60C and should be 1.18 rather than 1.225. Likely not matter one jot in the real world but whoever did the sum would be marked down in the 2396! I do remember the time when the calculation for 70C copper was to 115C (the average of 70 and 160).

In any event, the 1.1 uplift is a strange consideration. Is it saying that the resistance of the wire armour is increased by the magnetic effect of itself?

• In any event, the 1.1 uplift is a strange consideration. Is it saying that the resistance of the wire armour is increased by the magnetic effect of itself?

That's one way of looking at it, but I think it's possibly a simplification of a more complex situation, and what we have is an approximation at experimental results?

the correction should be to 60C and should be 1.18 rather than 1.225

Interesting ... we're using an internal conductor temperature of flexible cables for fixed installation wiring now? The armour is inside the cable, so unless you know the thermal properties of both the filler and the sheath, we just don't know ... BUT by the same token, perhaps we ought to consider the uninsulated cpc of T&E in the same way?  ... we really don't know its temperature(even if it was a full load), so the ONLY option is to assume it's the same as the current-carrying conductors ?

Made worse by not knowing the installation conditions, and distribution of temperature around a concentric cable ... unless the lay-up is helical with a parallel-axial wavelength suitably short with the length of run, again it's another unknown !

Pedantry is one thing, but sadly ...

• we are all struggling with algebra here ;-)

The current in the armour must go up with increasing external CPC reistance, and in the report it does.

Note that there  is a problem with this graphic, all the lower bracket is squared, and the root should extend over both the bracket and the 0.2^2

And the current in the external CPC must go up with rising armour resistance and indeed in the report it does though for reasons unclear the reactance does not appear in this one.

And these are still approximations that are only a "reasonable"  fit to 85mm cable.

Mike

Mike

• we are all struggling with algebra here ;-)

The current in the armour must go up with increasing external CPC reistance, and in the report it does.

Note that there  is a problem with this graphic, all the lower bracket is squared, and the root should extend over both the bracket and the 0.2^2

And the current in the external CPC must go up with rising armour resistance and indeed in the report it does though for reasons unclear the reactance does not appear in this one.

And these are still approximations that are only a "reasonable"  fit to 85mm cable.

Mike

Mike

Children
• Mike your equation for Ia is taken from page 26 of the ERA Report (in the format I have) which is used to produce Table 13. This differs from the final equation proposed in the ERA Report (page 29) in that the 0.7 factor becomes 1.1. This final equation has apparently been based on work in IET GN6 rather than the experimental data. The square root symbol extends as you said.

Regards

Geoff Blackwell

• taken from page 26 of the ERA Report

Indeed it is..

Arghh but it can't need to be both rounded up and also rounded down to be a better empirical 'fit' to the same results. If I was not working silly hours at the  moment I'd try and find the time to crash it through from their reported amplitudes and time delays because (at least) one of them must be wrong.

The other secondary lesson here says something about measuring with a micrometer something that gets tested by an axe.

What should be said is that the PSSC will be large, and it will not be quite what any of these formulae say, indeed even what we really mean by RMS current for an imulse waveform where no two half-cycles are quite the same is unclear.

Nine out of ten SWAs do not need a parallel CPC, unless you employ folk who cannot use a set of spanners and do not understand where not to paint and which side to put star washers.

Mike.

• I agree with most of that.

There are times that a parallel cpc can be needed such as bonding in some situations, disconnection times, etc – but, in general, they not required provided the cables are properly installed. I would add a copper connection to the earth terminal via a banjo washers as well as the other measures to ensure a good connection.

As to the micrometer and the axe, I definitely agree. We should bear in mind that cables come in discreet sizes so a calculation that indicate that an 82.345 mm2 cable is required are a waste of time – a quick calculation using a simple model might lead you to a 95 mm2 cable in one hit .

Regards

Geoff Blackwell

• As to the micrometer and the axe, I definitely agree. We should bear in mind that cables come in discreet sizes so a calculation that indicate that an 82.345 mm2 cable is required are a waste of time – a quick calculation using a simple model might lead you to a 95 mm2 cable in one hit .

Yes but if you are a contractor and a simple model says the cable should be 97.65mm2 and requires you to use 120mm2, you might be grateful for the micrometer approach that confirms you can use 95.

• No you just tweak one of the parameters so the answer becomes 95!!!!!!!!!!