ProMbrooke:

Zoomup:

ProMbrooke:

Zoomup:

ProMbrooke:

gkenyon:

When PME was introduced, we were still in the "post-war" period.


Now we are in a position where, in existing urban areas, PME is here to stay.


I understand some DNOs are offering TN-S for new-build - of course, there is a slight cost increase for the extra core.


With TN-S, you only need one failure after broken PE, on circuits where RCDs are not used. Where RCDs are used, I thought you said you weren't happy with reliance on the RCD?


Your point regarding earthing is very valid ... for all system types. The impact on the effectiveness of main bonding due to the change to plastic service pipes can't be underestimated. I have measured the effective combined earth electrode resistance of my water supply pipe, which is still lead from the street, and it's well under 4 ohms.


How should we deal with that? Germany insist on foundation earth electrodes, which achieve a similar result ... but that idea keeps being shot down here in the UK. I'd like to see some alternative approaches considered. Don't forget, the loss of plastic service piping is not a fault of the DNO, it's a result of health & safety (replacement of corroding metal gas mains for plastic) and public health (replacement of iron and lead water pipes for plastic) and far outside the DNO's control.

Two failures for TN-S without RCD, Three failures with RCD. RCDs aren't 100% reliable, hence why I advocate a foundation of low loop impedance and rapid disconnection. 

So what happens when I am outside in the garden and I get a shock on an old non R.C.D. supplied socket, due to a fault with an appliance or appliance flex, if I am in contact with 240 Volts mains. Let's say 5 Amps is flowing through my body. Will the 13 Amp plug fuse blow and save my life?


Z.

.

Thats where CPCs come in, in that they would cause the fuse or MCB to open in 0.2 seconds or less.

Not if the cable is just two core they wouldn't. Or if the appliance is all insulated=Class 2.


Z.

Right, where you would have a double insulated tool, which is not capable of becoming live ie made of plastic. 

And if the appliance flex is damaged but unnoticed. Perhaps insulation chewed through by mice or the appliance mechanically damaged by poor rough handling. If I come into contact with a live core or part then there is no protection by a C.P.C.


Z.

ProMbrooke:

I remember on another forum someone posting a TN-S overhead earthing supply from the 40s or somewhere about there. I feel like back then people understood electrical theory to a great depth.

Overhead TN-S:  I'm sure I remember references to this being used in the UK at some time. Just now, the only one I can find is in the work by Gosland, mentioned below: and there it's only an option, called "Direct Earthing to Separate Earth-Wire, Singly Earthed", without a claim that it was used in practice. I first thought it was mentioned in an old book, ~1929: Rural Electrification .. I don't see it there now, but the book should be fun for someone to glance through anyway, with its many diagrams. It was written in England by a Swedish engineer who worked there for a few years on electrification projects.


Overhead TN-C-S was, to the best of my knowledge, the original use of 'PME' in the UK. It was studied and trialled by ERA (electricity research association) in the late 1930s. Voltage-operated ELCBs with TT ('normal earthing') were also discussed. Both were seen as options for safety in rural supply, if one didn't want a further conductor.  Fact-finding missions (probably not by that name) were made to German utilities and factories. The German utilities had done some chopping and changing between systems. As was mentioned in a thread here some time ago, they tend in Germany even now to want a particular LV network to have one or the other type of customer earthing, not mixed TT/TN*. Concerns with 'PME' in the UK were not just the needless-to-say ones, but also the expectation of some normal load currents straying through the ground ... concern for the sake of telephones and telegraphs. On the factory side, one of the 1930s works mentions a German case with one voltage-operated ELCB per machine-tool, where operators were seen rigidly following the instruction of pressing to test before starting their work.


TN-C-S in urban (cabled) networks was - as others have mentioned - presumably post-WWII, when new cable designs were used, the sheaths of earlier cables having provided a free extra conductor. Trusty lead-sheathed cables, with their sheath and possible armour in contact with soil (any covering being water-permeable) also gave good earthing of the network's neutral-point by forming basically a network of earthing electrodes. The change to plastic-sheathed cables has changed the earth-potential rise on LV networks in the event of faults from the primary side of the transformer, which I know has concerned some countries. Dangers coming in from the higher voltages are a situation for which TT typically has an advantage over even TN-S, as it makes the installation earthing independent - at least until insulation breakdown or surge-protection operation.  Thinking the other way (about how TN-S isn't always the ideal in every way), safety to 230 V appliances against getting an extreme voltage is an advantage of TN-C-S with multiple earthing in contrast to TN-S.

 

I sympathise with the point about the depth of understanding in the old days. The main concerns of earthing methods have been discussed and analysed for over a century. Situations have changed (and so has terminology!), so old works aren't necessarily very useful in their conclusions; but there's a lot of old explanation that people nowadays who come fresh to the problems would find useful. There are several good old works I'd like to share, including the ERA ones, but can't as they're still in copyright, or at least the official scans are claimed to be. Anyone with IET library credits to be spent could look here for the 1937 and 1941 ERA papers. More impressive is the effort by Gosland (1950) to do a quantitative comparison of all the dangers of all the options. Inevitably it is built on a raft of assumptions, but it has some good discussion and plots. I'm sure I've pointed it out here before, sometime, mentioning the author's response to a questioner who noted the shakiness of assumed numbers ("... Nevertheless, any engineer attempting evaluation of the relative merits of methods of earthing must have quantities of this kind at the back of his mind, and there seems to be advantage in stating explicit figures, so that the foundations of opinion may be exhibited and discussed.").  Doubtless very good analyses of many of the issues were done in Germany or its neighbours, well before the fact-finding studies of the 1930s ERA reports.  One advantage in the old days was that people had to choose their analyses carefully and do them by hand, instead of putting a very detailed test-system into a simulation program and hoping to get a general answer out. Understanding was needed from the start.


Regarding the mentioned foundation-earthing in Germany (or separate electrodes): this can be of some use against a PEN-fault in any case, since domestic loads most of the time are small. But it's particularly useful with 3-phase supply, where the typical neutral current is only a few amps instead of tens. Major loads such as cookers, water heaters etc traditionally don't even have a neutral connection in countries that have 3-phase installations as normal. Then the neutral really is just a balancer for a few small loads, meaning that a super-good earthing isn't necessary to control its voltage if the supply neutral gets a fault. In the UK situation it wouldn't take many loads together to reach a dangerous neutral voltage when the current has to pass through a modest electrode and normal soil.

 

gkenyon:

ProMbrooke:


However, when all is said and done, I think table 41.1 needs to be re-visited again. 

But is that based on the fact that RCDs can't be trusted?


Table 41.1 is dry condition only.


Other than reliance on RCDs to achieve 40 ms disconnection time for additional protection, is IEC 60364 deficient in any other respect?

If all disconnection times are reduced to 0.2 s, what do you propose we do about circuits > 63 A (sub-mains etc.) that currently have disconnection times of 2 s and 5 s (and do NOT align in any way, shape or form to IEC 60479)? How will we achieve disconnection times and selectivity for these circuits?


If Table 41.1 is revisited for these reasons, other parts must be also ... would be good to hear your recommendations and reasoning on these other circuits?

In my opinion, an RCD can fail, and thus are just one layer of the onion.


There is also the fact RCDs are 30ma devices, where the no let go current starts at 10ma. It is possible (in theory) to end up frozen to the source, yet not trip the RCD. For this reason US GFCIs trip around 6ma.


In any case 6ma GFCI, or a even 30ma RCD, may trip large equipment due to leakage current. Even if/where steady state leakage current is under 6 or 30ma, Xc=1/2pifC can cause tripping from a sharp rising edge or spike in the sine wave. For this reason medical equipment, freezers, refrigerators, process equipment and the like should not be protected by an RCD due to the possible inadvertent tripping causing a greater risk such as food poisoning. Rather BS7671 should offer either an exception where the CPC is not likely to become compromised (like a home refrigerator that is typically rarely moved) or in the case of commercial kitchens and factories allow for other means such as an earth proving unit. The earth proving unit will not power the equipment unless the CPC is intact, and of course a 0.2 or 0.4 second disconnection time will take care of the rest.


US GFCIs tripping on AC units: 


   https://www.mikeholt.com/newsletters.php?action=display&letterID=2339  



There are many aspects of IEC 60364 that I would like to changed, clarified or added to. Or in the case of AFDDs have their mention tossed out altogether. However I think going into more detail on those other aspects would complicate this thread.


I do not believe all disconnection times should be reduced to 0.2 seconds, just wet locations. 0.4 can remain. 5 seconds for circuits over 63 amps, 10 seconds for circuits over 400 amps 15 seconds for circuits 1,200 amps and over. 30-60 seconds for select DNO cables.


Fault on larger circuits cause voltage drop on the output terminals of the transformer, resulting in lower touch voltage, and as such disconnection times can be longer. Typical transformer sizes, DNO runs, sub/final circuits and MCB/fuse trip times should be evaluated across the globe in order to establish reasonable values for each country.

mapj1:

Do we want to save all of the people all of the time, or is it more like car safety where there is a balance  to strike with what is reasonable versus inconvenience like trips firing when they should not ?

Perhaps  is worth noting that the DNOs work on live LV most of the time, including in holes full of water, and their networks break all sorts of BS7671 rules about earths, neutrals, discrimination, disconnection times, but with few incidents, suggesting BS7671 is over cautious.

Faults on large cables are rarer, as they are well, larger, and more obvious and also folk will not normally be holding onto anything fed directly by it, but rather things fed via a final circuit more finely protected.  (not true if you are the sort in the habit of opening the bus bar chamber to obtain a  temporary power of opportunity with some wingnuts but that is your own look-out - bring your own gloves....)


I agree it would be good to know what the assumptions are, for there surely are some, but the accident record suggests they are not that unreasonable.

Mike.

Well, I see videos of fualts breaking rather quickly on UK networks, where on other parts of the world they take their jolly time. UK DNOs know their craft in comparison.


I think we can save people all of the time, with less when all is summed than is being asked of right now...


 

gkenyon:

Mike,


I think these assumptions are roughly well published, at least historically, and "tweaks" from there are seen as "improvements", such as additional protection by RCD.


I do have a minor concern that for new installations there's definitely no such thing as an equipotential zone any more when you move outside steel-framed / steel-clad buildings, but if that's the case there are only other exposed-conductive-parts to touch anyway, and they are all connected to MET.


Outdoors, it remains as it always was in that respect, save for RCDs for socket-outlets and portable equipment.

I'm still not seeing profound elucidation in the IEC technical reports, or what testing exactly results in the inferences made. There is also the fact what little there is in comparison, is hidden behind exceptionally high costs which is only setting humanity backwards.


But worst of all, I'm seeing manufacturer driven influence and bias. Mandating RCDs and metal consumer units where there is also the option of earth proving units and thermarestor strips is not the intent of the regs. They should state what they want done, not what product should go about doing it. 


Regarding bonding, I am on the side of the IEC in that bonding just makes for a larger zone of energized metal. ADS should be the central focus IMO. Not earthing and bonding which are dated, misunderstood concepts rooted in ignorance.   

There are many aspects of IEC 60364 that I would like to changed, clarified or added to. Or in the case of AFDDs have their mention tossed out altogether. However I think going into more detail on those other aspects would complicate this thread.

If you are in the UK, then:

• It is possible for interested professionals to get involved, or indeed for anyone to submit comments and suggestions: https://standardsdevelopment.bsigroup.com/committees/50001574

• If you are an IET Member, there is another opportunity, see:
  
  • https://electrical.theiet.org/bs-7671/about-bs-7671/
  
  
• https://electrical.theiet.org/get-involved/

If you are not in the UK, then your national or regional standards organisations are probably glad of the help.

 

ADS should be the central focus IMO. Not earthing and bonding

AJJewsbury:

ADS should be the central focus IMO. Not earthing and bonding

I do have some sympathy for the idea that in a TN-S system, main bonding provides a limited and undefined benefit for faults in final circuits.


But what about mixed disconnection times?  A fault on a sub-main will likely take an extended time to disconnect (to ensure discrimination with downstream devices if nothing else) - perhaps up to 5s - the voltage on the fault is then usually imposed on the c.p.c.s of downstream circuits - including circuits that would themselves require 0.4s (or 0.2s) disconnection times. Without bonding at the local DB anyone in contact with exposed-conductive-parts would be exposed to the full fault voltage for a long duration compared with extraneous-conductive-parts or the general mass of the earth. Bonding certainly isn't a cure-all, but it would seem to weight the dice in our favour.


Also in TN-C-S/PME systems bonding helps to protect small c.p.c.s from carrying diverted N currents - so there are other risks to consider, not just shock.


Then in TT systems, bonding provides some backup should an RCD used for ADS fail.


  - Andy.

Higher rated circuits has a much lower R1+R2, producing voltage drop on the transformer and in turn a lower touch voltage whereby disconnection can be longer without worry of physiological harm.