table 41.1 max. disconn. times and the extra notes

"And the answer is...."; Nope i  dont 'get it' @mapj


I cant see in what you have said that suggests why the relaxation (use TN times instead of the more stringent TT times) cant be made for circuits not covered by 411.3.2.2.


What is it between a 32A socket circuit that can be relaxed and for one example (there might be others), a 40A distrib  circuit that can not be relaxed.


Perhaps I am at the limit of my now lazy and too simplistic mind I fear :-)

I'm not deliberately trying to be obtuse or incomprehensible, and neither are the reg writers. However at the risk of muddying the water, I'll try a different attack.


In general  such circuits as this applies to may need to be also be protected by RCD if they supply sockets but  this is for additional protection for high resistance events between live and earth  such as a person between a live wire and terra-firma.

This table 41.1 is not really about that, but about faults inside equipment from a live conductor to the CPC - such as a failure of a winding on a transformer or motor in class I kit, so the windings connect to the earthed core.


Let us only consider 230V AC supplies for now, and table 41.1 collapses to two figures 0.2 seconds for TT, where the maximum shock voltage  on earthed metalwork during fault is assumed to be 230, and 0.4 seconds for TN, where the peak voltage that might appear on exposed metal is perhaps half to two thirds of the supply (for reduced CPC T and E).

Here there is a relaxation on TT ciruts where the likley shock voltage is reduced.

Distribution circuits are something else, because unlike fixed low current or plug in equipment there is an assumption that no-one is holding the CPC during the time that the ADS operates and the fault is cleared.

The figures of 5 seconds or 1 second translate to being some way up the thermal part of the MCB curve,  and are more about protecting things being likely to overheat than shock protection, and at the same time not removing the supply to a bunch of circuits due to a fault at the far end of one of them.

You may wonder what happens to folk holding a portable appliance on a final circuit supplied by such a distribution circuit, but that is not considered. If it was you'd need 0.2 or 0. 4 seconds all the way back to the origin, and there would be no time discrimination possible,

So why 1 second for TT and not 5 - probably again due to the higher exposed voltage, and the greater exposure, in that assuming the electrode dominates, and not the wiring, all the metal work comes live

In contrast in a TN system, generally the cables get thinner towards the fault and therefore, much as the voltage on the live is sloping down towards the fault the voltage on the CPC is sloping upwards,

So would it be safe to relax the 1 second to 5 seconds for TT systems with a very low Zs and or a very light load ? not really, as there is still the larger area of exposed live stuff to consider.


Actually 5 seconds is a really long time to be standing against an almost unlimited fault current on any system and in practice you'd not want that very often, even if it is permitted.

@mapj "'m not deliberately trying to be obtuse or incomprehensible, and neither are the reg writers..."


I would want you to know that there is no reason to presume anyone would be thinking you (or the Reg writers) would be - same goes to the many others on here prepared to generously assist on a range of questions - I didn't and wouldn't presume that and its a shame my words gave that impression.


Thank you for the additional comments. What you state makes sense...and dare I say now, obviously ! ;-)


I was stuck on the thought that where a low impedance existed to allow an OPD to be used, then it would likely clear other circuit faults just as rapidly and did not give thought to the 411.3.2.4 retained for protecting against risky longer TN clearances being designed in.

Glad to be of help -  and do not worry,  I was only trying to warn that this might not actually help either - after all I had understood my own earlier post as well ! but it is hard to guess how others are looking at the same problem.

There is a large random element about the exact time limits for a lot of this - it is partly because we kind of like a rule of 3 for discrimination between successive layers of protection or perhaps 2 and bit if we are in a really tight spot.

So in the dim and distant past this was decided that  the front line may be a 13A fuse, backed by a 30 backed by 60, 80  or 100 and a  few hundred at the substation.

When trips  with both coils and bimetallic strips came along we had the option of a fast and slow part, and  got a bit more scientific with the time to break curves  - the bimetallic part more or less matching the fuse curves of old, handling overloads and the near-instant magnetic trip for dead shorts reducing the stress and permitting smaller contacts and thinner metals in a few places.


The only times that are not simply set by a committee furtively chewing their metaphorical pencils, are the ones set by the human heart beat period and the currents that can cause fibrillation - namely the shock protection table we have been discussing. The slower times are all about trying to avoid unnecessary disconnection of perfectly satisfactory bits of the installation, and at the same time not have too long a fault duration to cause damage, either by overheat or by voltage rises or depressions on the rest of the network. They could probably just  as well have said 4.1 seconds and 1.83 second,but round numbers are easier to use.

There is a lot of the spirit of "measure with micrometer, mark in chalk, cut with axe"  when folk agonise about small changes to Zs or breaking times.  Being just a marginal pass or just a marginal fail is neither that useful or that serious - being a clear pass is what is needed, but that is not compatible with a simple set of instructions.

"measure with micrometer, mark in chalk, cut with axe"

 I like that one