Copper equivalence

[EFLImpudence]

Leaving aside the question of CPCs in another thread.

As far as bonding is concerned it is generally stated/accepted that the swa has to be approximately 8.5 times greater as it must have the equivalence of copper.
This is, though, based solely on the resistivity comparison of copper and the armour. Steel has varying values for resistivity so I assume someone has determined this value to be correct for the armour type.

However, as pointed out in the other thread, the 'k' values of copper and steel are 115 and 51 respectively.
These values have a ratio of 2.255:1
This value corresponds to the charts published for csa comparison (K2/K1 x swa csa) which are discounted as not adequate for the bonding requirement.

Would it therefore not be correct to use this figure, 2.255, when determining the required csa of the armour when used as bonding as the actual resistance is not a consideration?

Or, on the other hand:

As the bonding conductor size is to allow for the heating of the conductor that may be encountered and this is dependent on the resistance, should the steel not be required to be 8.5 x 2.255 = 19.17 times that of copper?

 

[JohnW2]

Leaving aside the question of CPCs in another thread. ... As far as bonding is concerned it is generally stated/accepted that the swa has to be approximately 8.5 times greater as it must have the equivalence of copper.
This is, though, based solely on the resistivity comparison of copper and the armour. Steel has varying values for resistivity so I assume someone has determined this value to be correct for the armour type.

Yep, I think that ~8.5 is generally taken as the relative resistivity of 'average steels' as compared with copper.,

However, as pointed out in the other thread, the 'k' values of copper and steel are 115 and 51 respectively. These values have a ratio of 2.255:1 ... This value corresponds to the charts published for csa comparison (K2/K1 x swa csa) which are discounted as not adequate for the bonding requirement.

Not quite. Discounted as not satisfying Table 54.7 - which relates to any protective conductor (i.e. CPCs as well as bonding conductors). However, as someone pointed out (by implication) in the other thread, only lazy people use Table 54.7. Non-lazy ones undertake an adiabatic calculation and, almost invariably, get a much smaller answer as a reward for their non-laziness!

Would it therefore not be correct to use this figure, 2.255, when determining the required csa of the armour when used as bonding as the actual resistance is not a consideration?

That is effectively what you would be doing (for any sort of protective conductor) if you undertook an adiabatic calculation using K=51. However, if you tried to do an adiabatic calculation for a bonding conductor, I'm not sure what "I" and "t" you would use. Indeed, come to think about it, I'm not even sure that an adiabatic calculation would be appropriate - since, in the absence of any protective device guaranteed to rapidly break the current, it would be a continuous current - not an adiabatic situation at all! Indeed, if one assumed that the current might be continuous, one would presumably have to use a cable with a high enough CCC per the usual tables (Appendix 4) for the perceived maximum current - which would presumably result in an extremely large bonding conductor!!

Or, on the other hand: ... As the bonding conductor size is to allow for the heating of the conductor that may be encountered and this is dependent on the resistance, should the steel not be required to be 8.5 x 2.255 = 19.17 times that of copper?

This is getting complicated As above, I'm not sure what current (or duration) one would assume. However, if it were (at least potentially) a very high current (maybe due to one of the infamous "lost TN-C-S neutrals"!), that would imply that the external parts of the current loop was of very low impedance. If one had a 'high resistance' steel bonding conductor, that would actually reduce, perhaps considerably, the current which could flow (hence heat which would generated). Like putting two lamps in series across the supply - the increased resistance would considerably reduce the total current, thus even more considerably reducing the amount of light and heat generated.

Having said all that, although I realise it's not a 'regulation', what about the "0.05Ω maximum" guidance for main bonding conductors?

Kind Regards, John

 

[EFLImpudence]

Having said all that, although I realise it's not a 'regulation', what about the "0.05Ω maximum" guidance for main bonding conductors?

My view on that is that it is just a figure quoted which is acceptable as negligible impedance ensuring good connection of the conductor.

I accept that that doesn't really make much sense but, as you say, there is nowhere any mention of the impedance of a bonding conductor.

 

[JohnW2]

Having said all that, although I realise it's not a 'regulation', what about the "0.05Ω maximum" guidance for main bonding conductors?

My view on that is that it is just a figure quoted which is acceptable as negligible impedance ensuring good connection of the conductor. ... I accept that that doesn't really make much sense but, as you say, there is nowhere any mention of the impedance of a bonding conductor.

Indeed - although there obviously are minimum CSA figures for bonding conductors, so if one assumes a 'maximum credible length', that implies a certain maximum impedance.

Kind Regards, John

 

[EFLImpudence]

Not quite. Discounted as not satisfying Table 54.7 - which relates to any protective conductor (i.e. CPCs as well as bonding conductors). However, as someone pointed out (by implication) in the other thread, only lazy people use Table 54.7. Non-lazy ones undertake an adiabatic calculation and, almost invariably, get a much smaller answer as a reward for their non-laziness!

Accepted but what about the blanket requirement for PME of 10mm² for the main bonding conductor?

Surely, an armour equivalent of 85mm² is unrealistic and totally unnecessary.
So, would 22.55mm² not be acceptable?

 

[JohnW2]

Not quite. Discounted as not satisfying Table 54.7 - which relates to any protective conductor (i.e. CPCs as well as bonding conductors). However, as someone pointed out (by implication) in the other thread, only lazy people use Table 54.7. Non-lazy ones undertake an adiabatic calculation and, almost invariably, get a much smaller answer as a reward for their non-laziness!

Accepted but what about the blanket requirement for PME of 10mm² for the main bonding conductor? ... Surely, an armour equivalent of 85mm² is unrealistic and totally unnecessary.

Maybe unrealistic, and possibly 'unnecessary', but AFAICS, there is no doubt that that is what note to Table 54.8 requires ("actually says"!) - a conductor of equivalent conductance (i.e. equivalent resistivity) to that of a 10mm² copper conductor - i.e., as you say, at least 85mm².

So, would 22.55mm² not be acceptable?

As above, seemingly not acceptable to the regs, which would require a minimum of 85mm² if steel.

Kind Regards, John

 

[EFLImpudence]

Can't argue with what the regulations actually say.

Indeed - although there obviously are minimum CSA figures for bonding conductors, so if one assumes a 'maximum credible length', that implies a certain maximum impedance.

Shouldn't that be 'maximum credible impedance'?

Would 0.1Ω be disasterous?

 

[securespark]

EFLI...I can see what you're saying.

I shall stick my neck out and say that I for one thinks that 85mm² is OTT and reckon that 25mm² would be adequate.

Another case of the regs over-engineering when it comes to cable sizes?

Now I am about to dive behind the sofa...Tara!!

 

[EFLImpudence]

Thanks ss.

Does it depend on what is meant by 'the same conductance'?

Whilst there is a standard definition of 'conductance', presumably a 10mm² copper conductor of negligible impedance is required so that it can handle 60+A without damage should the need arise.

What csa of steel is really necessary to achieve the same?

 

[JohnW2]

Can't argue with what the regulations actually say.

Quite, and some people seem to put an awful lot oweight on what the regs "actually say"!

Indeed - although there obviously are minimum CSA figures for bonding conductors, so if one assumes a 'maximum credible length', that implies a certain maximum impedance.

Shouldn't that be 'maximum credible impedance'?

Yes, that's a fair comment.

Would 0.1Ω be disasterous?

I wouldn't have thought so - although, I suppose, in one of bernard's "hundreds of amps" scenarios, 0.1Ω would translate to "tens of volts" between exposed-c-ps and true earth (after parallel-path CPCs had melted!).

Kind Regards, John

 

[JohnW2]

I shall stick my neck out and say that I for one thinks that 85mm² is OTT and reckon that 25mm² would be adequate.

Intuitively, I agree, and I imagine that most other people would probably say the same. However, we are talking about what the regs "actually say"!

KInd Regards, John

 

[JohnW2]

Does it depend on what is meant by 'the same conductance'?

As you go on to say, there is a very clear and universally accepted definition of 'conductance', and we all know what 'same' means!

Whilst there is a standard definition of 'conductance', presumably a 10mm² copper conductor of negligible impedance is required so that it can handle 60+A without damage should the need arise.

I'm not sure I understand what you are saying. What do you mean by "a 10mm copper conductor of negligible impedance"? A 10mm² copper conductor of a known length will have a known impedance - what do you regard as 'negligible'. Whatever, it's not really the impedance (negligible or otherwise), per se, that determines what current it can handle, is it?

What csa of steel is really necessary to achieve the same?

If we were definitely talking about an adiabatic situation (high current for less than ~5 secs) (which we probably aren't), then it presumably would need to be 22.5mm² (based on k1/k2). If we have to assume (which I think we probably do) that the current could be 'continuous', then I haven't a clue, since I don't have any CCC tables for steel conductor cables - do you? What one needs to know is the 'continuous' CCC for a steel conductor that corresponds to the CCC of 10mm² copper conductor, with the same 'installation method'.

Kind Regards, John

 

[ban-all-sheds]

Whatever, it's not really the impedance (negligible or otherwise), per se, that determines what current it can handle, is it?

Actually, it very much is.

There are other factors, but they are all related to how it behaves as it heats up, and it is the fact that it has impedance which means it heats up. If it had zero impedance it could carry an infinite current.

 

[JohnW2]

Whatever, it's not really the impedance (negligible or otherwise), per se, that determines what current it can handle, is it?

Actually, it very much is. There are other factors, but they are all related to how it behaves as it heats up, and it is the fact that it has impedance which means it heats up. If it had zero impedance it could carry an infinite current.

That's all true, and I can see that I expressed myself fairly badly.

We are not talking about a situation of a 'constant current source'. We are talking about an (incredibly rare) situation in which a very high current is flowing through a bonding conductor and, in that situation (and particularly if the conductor is made of a material of relatively high resistivity) the resistance of the bonding conductor is likely to represent a substantial proportion of the total impedance of the entire current path. In that situation, increasing the impedance of the bonding conductor is likely to decrease the current - hence increasing, rather than decreasing, its ability to handle the situation. Just as a bonding conductor of zero impedance could carry an infinite current, a bonding conductor of infinite impedance would never heat up, or come to any harm, at all.

Kind Regards, John

 

[EFLImpudence]

Looking at it from the other direction, why is the k value for steel 51 and not 13 - 115/8.5?

If this is because of other factors relating to the difference between steel and copper, then why cannot the 2.255 value be used for sizing bonding conductors as well as CPCs - or vice versa.