In a recent thread ...
Unlike the situation with TN-C-S, in a TT (or TN-S) installation there is no risk that, under certain fault conditions, the combined neutral current of several installations might try to travel through one's main bonding conductor. Accordingly, the only situation in which high currents might flow through the bonding conductor is as a result of an "L-E" fault within the installation, and even that only if the bonded extraneous-c-p represents a low resistance/impedance path to true earth.
Of course, if (as 'required') everything in a TT installation is provided with fault protection with RCDs, then an RCD should clear any L-E fault very rapidly, so this discussion becomes essentially moot. The same is essentially true of a TN-S installation, provided that all circuits have (as 'required') adequate fault protection (from OPDs and/or RCDs).
When there is an "L-E" fault, the fault current will be shared between the installation's earth and the bonded extraneous-c-p. Assuming the bonded extraneous-c-p has a fairly low impedance (much lower than that of the TT electrode) to true earth, that means that (until the fault is cleared) in a TT installation, most of the fault current will flow through the bonding conductor, whereas in a TN-S installation the proportion going through the bonding conductor will depend on the relative impedances of the two paths to earth.
Hence, since the problems specific to TN-C-S do not exist, and the only high currents that can flow through the bonding conductor with TT or TN-S are those arising from faults within the installation, then in terms of the ability to 'cope with' those potentially high currents, the main bonding conductor is in no different a position from the CPC of the circuit with a fault.
In other words, in terms of that consideration, if the CPC of the circuit with the highest PEFC is 'adequate' (adiabatically), then there is no apparent electrical need for the main bonding conductor to have a larger CSA than that CPC - which in practice means, electrically speaking, a much smaller bonding conductor than 6mm² would invariably be 'adequate.
So, from that consideration (ability to cope with fault current) the 6mm² minimum for main bonding conductors in TT and TN-S installations seems to be essentially arbitrary, not based on 'electrical need'. Nor can I really think of any other 'considerations'. They might have been concerned about possible voltage drops along the length of the bonding conductor (possibly leading to unacceptable PDs), but since there CSA requirements do depend on conductor length, that does not seem to be part of their thinking.
Are there any flaws in my thinking? I suppose a (very improbable) L-E fault upstream of final circuits might possibly result in higher fault currents in a bonding conductor than would be possible with a fault on any of the final circuits?
Kind Regards, John
This comment from @Adam_151 reminding me that I have never really been sure about the thinking behind requirements for main bonding conductors in a TT (or, come to that, TN-S) installation, the situation seemingly being that the 6mm² minimum for such conductors is essentially arbitrary (and electrically 'OTT'). However, my thinking may be flawed!So what EICR code would you give to undersized bonding?, personally I'd probably look towards a C3 for the 6mm main bonding on a domestic TNCS (remember though that if TT, it might not be undersized at all) ....
Unlike the situation with TN-C-S, in a TT (or TN-S) installation there is no risk that, under certain fault conditions, the combined neutral current of several installations might try to travel through one's main bonding conductor. Accordingly, the only situation in which high currents might flow through the bonding conductor is as a result of an "L-E" fault within the installation, and even that only if the bonded extraneous-c-p represents a low resistance/impedance path to true earth.
Of course, if (as 'required') everything in a TT installation is provided with fault protection with RCDs, then an RCD should clear any L-E fault very rapidly, so this discussion becomes essentially moot. The same is essentially true of a TN-S installation, provided that all circuits have (as 'required') adequate fault protection (from OPDs and/or RCDs).
When there is an "L-E" fault, the fault current will be shared between the installation's earth and the bonded extraneous-c-p. Assuming the bonded extraneous-c-p has a fairly low impedance (much lower than that of the TT electrode) to true earth, that means that (until the fault is cleared) in a TT installation, most of the fault current will flow through the bonding conductor, whereas in a TN-S installation the proportion going through the bonding conductor will depend on the relative impedances of the two paths to earth.
Hence, since the problems specific to TN-C-S do not exist, and the only high currents that can flow through the bonding conductor with TT or TN-S are those arising from faults within the installation, then in terms of the ability to 'cope with' those potentially high currents, the main bonding conductor is in no different a position from the CPC of the circuit with a fault.
In other words, in terms of that consideration, if the CPC of the circuit with the highest PEFC is 'adequate' (adiabatically), then there is no apparent electrical need for the main bonding conductor to have a larger CSA than that CPC - which in practice means, electrically speaking, a much smaller bonding conductor than 6mm² would invariably be 'adequate.
So, from that consideration (ability to cope with fault current) the 6mm² minimum for main bonding conductors in TT and TN-S installations seems to be essentially arbitrary, not based on 'electrical need'. Nor can I really think of any other 'considerations'. They might have been concerned about possible voltage drops along the length of the bonding conductor (possibly leading to unacceptable PDs), but since there CSA requirements do depend on conductor length, that does not seem to be part of their thinking.
Are there any flaws in my thinking? I suppose a (very improbable) L-E fault upstream of final circuits might possibly result in higher fault currents in a bonding conductor than would be possible with a fault on any of the final circuits?
Kind Regards, John