TNC

Are we really talking combined N & PE throughout the installation?


What are the circumstances of the buried cables? Are they run in ducts? Is the site such that they're likely to be disturbed/damaged?


Monitoring the non-N (PEN?) currents to try and detect an earth fault is an interesting idea. I suspect it's not going to be that reliable though. You need a pretty decent electrode to get as low as 15Ω to the general mass of the earth - even in pretty favourable soil conditions I get 30-50Ω from a 1.2m rod - to get 15Ω you'd likely need several metres of exposed conductor. Simple fork or spade or JCB damage to a buried cable I would guess result in a far higher resistance to Earth - and only a few tens of mA could be fatal for anyone touching the damage.


   - Andy.

Is there an NE linkanywhere - I'm wondering if could it be an example of PNB of the second kind, where only Ls and N come from the TX, and the N_E link is at or near the main switch gear A DNO like approach where the pre switch gear wiring is protected by 'neutralled' armour or equivalent covering may be possible.

Or possibly an entirely class II installation? (might explain the preference against armoured cables).

  - Andy.

First a caution - HV distribution is not my thing. But I have had a fair amount to do with gensets, large UPS and so on in the fairly recent past. Take what follows with a pinch of salt, it may contain errors.


I'd expect at the least an HV  earth electrode for the transformer core and chassis,  though that may well not connect to anything on the secondary side. That is to trip off the HV on a primary winding to core or casing sort of earth fault. (at the other end of the HV line, the effect of a jumbo RCD is done by looking at current transformers and balance, or if the supply is small, or old, then perhaps just expulsion fuses.

 

I'd also expect some part of the secondary side to be earthed, if not at the transformer then in an adjacent building. If this is not done the secondary voltage may well be the intended 230 or whatever between the wires,  but with both sides floating off at some fraction of the HT voltage relative to ground, via transformer inter-winding capacitance and perhaps insulation leakage, and you would never know until it was too late.

IT does not work in that sort of case without the equivalent of an earthed screen between the primary and secondary.


The LV earth may well have to be some distance from the HV one, if there are concerns about voltage rise (high HV side PSSC, and poor ground conductivity perhaps)

Un armoured transfomer tails are a risk if they are vulnerable to strke, but if the route is made very clear, or mechanically well protected, maybe not an immediate concern.


It is sometimes done that what is in effect an earth fault relay on the LV side is used to operate a breaker on the HV side. Others will be better placed to comment on the detail.


If adding earth fault protection, until you can measure the leakage in normal use, it is common to start sizing for 0.1% up to about 0.5% of the main current, perhaps  with twiddle options for a factor of 3 either way after commissioning. (which sort of figures with a 30mA RCD on a 30A final circuit,  compared  to  300mA  at the origin of the 100A per phase on a modest farm.)


How big is the transformer and what sort of vintage ? Funnier things were done in the dark ages.

Thanks guys


To answer a few questions at once, yes N&E really is combined in the private submains and then separated at switchgear immediately prior to outdoor equipment, so the neutral is well connected to earth at multiple points. It looks like someone's designed it as TNC(-S) with the separation occuring just before the loads.


The working assumption - not my scope to actually check, but it will be a stipulation - will be that ADS works assuming there's a zero impedance fault and that the risk of cable strike is adequately managed.


The questions really are whether there's more sensitive protection other than plain old overcurrent available for earth faults on TN-C, and similarly there's something reasonable that can be done to protect TN-S 3P+E unarmoured cable. The object in both cases being to limit damage to property (i.e. the installation).

Presumably the LV neutral is earthed at the transformer - and what you effectively have is a whole lot of external loads fed by the TNC part of a TN-C-S system


If so, there may well be earth fault protection in place by means of a CT on the neutral/earth cable connecting to the star point - ie it's looking at actual current flow returning through the ground compared to actual current flow going out - akin to an unrestricted earth fault relay arrangement  - this may also include a "frame earthing" relay.


It's a pretty unusual set up, so possible MoD, Water industry or something not totally dissimilar where cable damage is highly unlikely within the site boundary.


Regards


OMS

OK, that is not as immedately dangerous  as it sounded at first. Clearly a broken neutral in the distribution would be very bad, as it is for the DNOs so a similar level of caution with inspections and double crimping of any PEN joints  and all the rest,  should be exercised.

If this was NZ or OZ it would be quite normal, it does mean only the final circuits after the N-E split can be RCD protected, and the submains cannot be.

If this was connected to the UK domestic LV  network and shared that substation with other users, s it would be a very  clear violation of the ESQR.

I appreciate it is a private network, and the combined cables are in the distribution and not diirect to loads, so  it may be OK.but  I consider it on thin ice however. Section  8 clause (4) A consumer shall not combine the neutral and protective functions in a single conductor in his consumer’s installation "

Is pretty solid  and
35.Any generator, distributor, supplier, or meter operator or any agent, contractor or sub-contractor of any of the foregoing who fails to comply with any provision of these Regulations which applies to him, any person who fails to comply with regulation 18(3), 21, 22 or 25(1) and any consumer who fails to comply with regulation 8(4) or 34(2) shall be liable on summary conviction to a fine not exceeding level 5 on the standard scale

Requires it to be take seriously. Unless there is a grant of exception by the secretary pf state, but I imagine they would have mentioned that if there was.

(Sorry would quote properly for replies but seem to have lost the Quote button)

OMS‍: Yes that's how I'm looking at it having mulled for a day or two. There's not currently a UREF relay in the existing design (we're replacing the main switchboard). Wouldn't a UREF suffer from neutral current diverted into parallel earth paths?  To be fair so would the sensitive earth fault relay I'm proposing on the main N-E link.


Cable damage is indeed hopefully unlikely given that routes are theoretically well documented on a controlled site but it's been poorly constructed so I'm erring on the side of caution.

mapj1‍ Thanks; the client has been informed in no uncertain terms by others already and I've added my voice to the calls. Hopefully they'll do something about it but it's not in my remit and the instruction I've recieved is to deal with what's currently there as well as allowing for a corrected system.

The neutral connection to the star point wouldn't go through the CT coil only the t connection to the electrode - so any current flow is balanced (ie doesn't matter which way it's flowing) - the CT would only be looking at the actual amps returning via the electrode and deciding from there whether it's a trip action - possibly on 2 stage (eg 250A for more than 1 second or 50A for more than 0.4 seconds


It is really the last ditch (hence unrestricted) - it doesn't matter where the fault is, if the set points on the relay are exceeded then it's going off - nothing subtle about it


You say it's not present though  - so entirely reliant on the "quality" of construction and ongoing maintenance


Regards


OMS