table 41.1 max. disconn. times and the extra notes

So when the earth rod resistance is really low, then an MCB or fuse will be as fast as an RCD.


Presumably the thinking is that during an L-E fault the touch voltage is not as high, with say a 5 ohm electrode as it would be with a 100 ohm electrode resistance, so a longer break time is OK.


To meet safety of life with a lower voltage is only applicable to the breaking times of half a heartbeat or so, where we can stand 100volts for almost twice as long than we can stand 200V for the same death wish, 1 second or 5 seconds are all about fire/ overheat so the exposed voltage during fault is not such an issue.

It’s not impossible to achieve the disconnection times with a Type B MCB in a TT installation for a few lights or a small appliance with an earth electrode with a Ra less than 11 ohms.


But I cannot really imagine I would omit a RCD.


Andy Betteridge.

Though I do see quite a few TT installations where electricians have omitted the RCD protection.

OK so 5 ohm rod for a really small circuit like one lighting circuit in a barn perhaps. To reliably give TN-s like rise of earth potential/ shock duration against  a 50 or 63 A breaker needs loop impedances perhaps a 10th of that - which is not all that likely to be met at the supply transformer end if it is a modest rural one ( a single pole pig could be could be on an LV electrode as high as 20 ohms, but probably no more than half that), let alone the resistance  any electrodes at the load end.

Note that if you do have a really low electrode resistance at the load end, better than at the substation, then while fault current is flowing, all other users see the neutral voltage move off- earth , and the faulty live to earth voltage is depressed, while (if present) the other 2 phase to ground voltages rise.

I'd very much agree, given the natural variation of ground resistances and so on, it is a brave fellow who omits RCDs in a TT system, or maybe one who has not realised the full significance of it being TT.


Perhaps they are saved from total failure by the fact that a lot of TT systems in buildings with bonds to the main water pipes etc, are actually neutral connected as someone on the same substation, and same water main, has PME or TNS. I fully appreciate this is not how it is supposed to work.

fortuitous & serendipity, springs to mind

The question is really about having a 0.4 second disconnection time rather than a 0.2 second.


A bit more consideration is required

Just in case it was not obvious, my post was not about using RCD or not :-)


It was the 'notes' (extract in first post) I was musing about; presumably made just in the [rare] occasion the TT earth is extremely low that then the TT times can be relaxed to TN times if bonding is in place - possibly impled too, to also be fine to relax where bonding not required - but where that relaxation is applicable, seemingly there is no provision to relax to TN times for other circuits not covered in 411.3.2.2  ...  I was wondering why.

And the answer is the lower touch voltage with a low impedance earth - during fault the earthed case of a hand held load rises to a voltage that is set by the relative resistances of the L and E paths forming a voltage divider. With a reduced CPC in T and E,  that voltage  may be about 2/3 of the 230V at the far end of a long circuit,  though  with a sub-main with an equal L and E cross -section it would be half the phase voltage. In reality somewhere in between is typical on a TN system.


On a high resistance TT system ALL the exposed metalwork on every thing using the same CPC and also the surface of the ground near the rod itself rises to nearly the full phase voltage during a fault, so clearing that fault quickly is much more important

If the Ze is really low then with larger fault currents we get some of our voltage division back..