... although, if I'm interpreting the regs correctly, 411.4.4 allows one to rely upon an RCD for fault protection in a TN system if the Zs is too high. That seems to make sense, since with a TT system, in the absence of parallel paths, one has no choice but to rely on RCDs for fault protection.The advantage is that you can have greater start up surges on the circuit. ( approx. 10x In as opposed to 5x In for a type B). The disadvantage is that your Zs has to be half the value of a type B for an mcb of the same value.
I think you are (interpreting the regs correctly) in respect of TT where there is no alternative.... although, if I'm interpreting the regs correctly, 411.4.4 allows one to rely upon an RCD for fault protection in a TN system if the Zs is too high. That seems to make sense, since with a TT system, in the absence of parallel paths, one has no choice but to rely on RCDs for fault protection.
That is assuming a TN-C-S supply so the earth loop at the head will be same as the neutral loop which has to be 0.35 ohms or better.
A slip of the 'pen'In theory the PFC is which ever is the higher line to earth or line to neutral
I will agree but is it so hard to use a meter?That is assuming a TN-C-S supply so the earth loop at the head will be same as the neutral loop which has to be 0.35 ohms or better.
It does not "have to" it "should be"!
A figure of up to 0.8 is acceptable as the DNO 100A fuse will still operate at that figure.
This figure was taken from a Supply Industry Specification, which decided on the figure of 0.35 as that is what would be expected where the neutral and earth were combined.
Unfortunately BS7671 jumped on this figure without fully explaining it so it has become slightly abused in it's use!
Indeed. As I said, it's obvioulsy what has to happen with TT, since there is no alternative.I think you are (interpreting the regs correctly) in respect of TT where there is no alternative.
Well, all that emphasising, underlining and capitalisation is yours, not that of BS7671. 531.3.1 merely talks of situations in which 411.4.5 "cannot be satisfied", without any such emphasis. I would say that it's anyone's guess as to how the authors intended "cannot be satisfied" to be interpreted ....However, with TN it should be emphasised that RCDs may only be used for additional protection and where Zs for the MCB cannot be satisfied i.e. IMPOSSIBLE- not just isn't satisfied. 531.3.1
... which was, as you might guess, going to be the basis of my next question! TT systems exist and are allowed - which presumably means that they are considered to be acceptably safe. So, if it's considered safe for L-E fault protection to be dependent upon RCDs in TT installations, why not also in TN ones? [see below - I may have answered my own question!]If that were the case, then all requirements for MCBs could be ignored merely by the inclusion of an RCD.
L-N faults are obviously a different matter, and no different for TT and TN systems. As you say, it's then R1+'RN' (i.e. usually 2*R1), plus the external L-N loop impedance which matters, and that has to be sufficiently low to achieve the required disconnection time.After all the R1 may be the reason the maximum Zs cannot be met and this with Rn may result in the MCB not operating under L - N fault conditions.
As in a previous thread it was, I think, agreed that your system was better protected either by the service pipes or your neighbours' systems.Indeed. As I said, it's obvioulsy what has to happen with TT, since there is no alternative.I think you are (interpreting the regs correctly) in respect of TT where there is no alternative.
I emphasised the words that mean you can't rely on RCDs and ignore the MCB requirements in TN systems.Well, all that emphasising, underlining and capitalisation is yours, not that of BS7671. 531.3.1 merely talks of situations in which 411.4.5 "cannot be satisfied", without any such emphasis. I would say that it's anyone's guess as to how the authors intended "cannot be satisfied" to be interpreted ....However, with TN it should be emphasised that RCDs may only be used for additional protection and where Zs for the MCB cannot be satisfied i.e. IMPOSSIBLE- not just isn't satisfied. 531.3.1
Perhaps it may be revised as I suppose it still applies to the single RCD protecting the whole TT system deemed to be acceptable because unavoidable yet not ideal.... which was, as you might guess, going to be the basis of my next question! TT systems exist and are allowed - which presumably means that they are considered to be acceptably safe. So, if it's considered safe for L-E fault protection to be dependent upon RCDs in TT installations, why not also in TN ones? [see below - I may have answered my own question!]If that were the case, then all requirements for MCBs could be ignored merely by the inclusion of an RCD.
As above, the MCB will clear the L - E fault in TN. Was that considered preferable to one fault blacking out the whole installation?L-N faults are obviously a different matter, and no different for TT and TN systems. As you say, it's then R1+'RN' (i.e. usually 2*R1), plus the external L-N loop impedance which matters, and that has to be sufficiently low to achieve the required disconnection time.After all the R1 may be the reason the maximum Zs cannot be met and this with Rn may result in the MCB not operating under L - N fault conditions.
However, as above, I think I've probably answered my own question. Logic suggests that (despite your emphasising), reliance on an RCD for disconnection in the case of an L-E fault ought to be as acceptable with TN systems as with TT ones.
More likely the other way around.However, that doesn't alter the requirements for disconnection with L-N faults - and if the R1 and 'RN' are low enough to satisfy that, they are pretty likely (although not guaranteed) to also result in a sufficiently low Zs for adequate L-E fault protection from the MCB.
As Eric mentioned, there is no provision for this to be recorded but it is taken into account in the PFC recorded. If the worse case figures satisfy then the system is adequately protected.If that it correct, then, if there is RCD protection in place, the requirement in relation to L-N fault protection really ought to be in terms of L-N loop impedance, not Zs/EFLI.
Indeed, and that illustrates what I regards as one of the major strangenesses/ inconsistences about the regs/GNs. I am not allowed to 'rely upon' service pipes as my TT earth (unless they are from a private supply) seemingly because of bthe risk that someone will creep around one dark night and turn them into plastic without my noticing. However, when it comes to testing my installation, it seems accepted and expected practice to measure Zs with all parallel paths in place. As a result, for the reasons you mention above, the measiured Zs throughout all the circuits of my installation is low enough to provide the required fault disconnection times with MCBs alone. However, if that hypothetical person with a lorry full of plastic pipe appeared one dark night, and worked very quietly, my Zs figures would suddenly all become about 75Ω without my realising.As in a previous thread it was, I think, agreed that your system was better protected either by the service pipes or your neighbours' systems.
Yes, I'm one of 'them', but ....Lately however some posters have commented that they would feel safer with a TT system rather than relying on the supply neutral.
I don't think so. That's essentially the situation I have (TT system plus the 'benefit' of TN-C-S via the neighbour), but I think that's probably the worst of both worlds - since my ('high resistance') TT earth would do nothing to protect my installation from a high voltage neutral if the neutral fault were in a bad place in the distribution cabling.I am not clear about this myself but I suppose the better would be both.
I still don't think that the intended meaning of "cannot be satisfied" is anything like as clear-cut as you seem to believe. I particular, I doubt that it was meant to mean "would be totally impossible to achieve" - given only that ZE is not so high that it makes it literally impossible to attain a low enough Zs, one can always get a low enough Zs, even if it means using conductors the size of drainpipes!I emphasised the words that mean you can't rely on RCDs and ignore the MCB requirements in TN systems.
I take this to be in the event of a fault in an existing system; the cause of which cannot be determined and renewing the circuit would be impracticable (perhaps temporarily). Obviously it would not apply to a new design which was badly engineered.
Who knows. However, that is less of an issue with more modern installations with multiple RCDs and/or RCBOs.As above, the MCB will clear the L - E fault in TN. Was that considered preferable to one fault blacking out the whole installation?
That's why "(although not guaranteed)" was thereMore likely the other way around.....if the R1 and 'RN' are low enough to satisfy that, they are pretty likely (although not guaranteed) to also result in a sufficiently low Zs for adequate L-E fault protection from the MCB.
It's fair enough to say that in relation to a TN system, but it is probably totally untrue of a TT system. With true TT, if Zs (i.e. EFLI) is all one measures, one has no idea whether the requirements for disconnection times for a L-N fault are satisfied or not - they certainly wouldn't be on the basis of the Zs measurements. Only an L-N loop impedance could demonstrate that.As Eric mentioned, there is no provision for this to be recorded but it is taken into account in the PFC recorded. If the worse case figures satisfy then the system is adequately protected.If that it correct, then, if there is RCD protection in place, the requirement in relation to L-N fault protection really ought to be in terms of L-N loop impedance, not Zs/EFLI.
Possibly, but I have yet to see any concrete evidence that RCDs are any less reliable than MCBs. As discussed previously, I strongly suspect that the reason for this belief may be that faulty RCDs are detected (by householders pressing the test button or electricians undertaking RCD tests). If we could test MCBs, we might well find at least as many that had 'failed' - but there is no practical/ safe/ non-pyrotechnic way of testing an MCB, at least, not at the level of fault currents! MCBs might even be less reliable, because they rarely operate, and don't even get 'exercised' by people (occasionally!) pressing a test button.A lot of posters express disquiet about the reliability of the RCD. Perhaps they (RCDs) are not regarded, by the powers that be, as safe enough yet to do as you are suggesting.
MCBs might even be less reliable, because they rarely operate, and don't even get 'exercised' by people (occasionally!) pressing a test button.
There is no ambiguity or imprecision in the meaning of "cannot" - look it up.I would say that it's anyone's guess as to how the authors intended "cannot be satisfied" to be interpreted ....
But it's not.With true TT, if Zs (i.e. EFLI) is all one measures,
L - N loop has to be measured to determine the PFC.one has no idea whether the requirements for disconnection times for a L-N fault are satisfied or not - they certainly wouldn't be on the basis of the Zs measurements. Only an L-N loop impedance could demonstrate that.
If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.
Select the supplier or trade you require, enter your location to begin your search.
Are you a trade or supplier? You can create your listing free at DIYnot Local