ELI pass point.

[ericmark]

I am told the DNO can set up multi-feeds with fuses drawn or inserted so they can isolate sections to work on them. Not in all cases, but we who don't work for the DNO will not know if the supply has changed.

So we are also told TN-C-S domestic is at least 0.35Ω and TN-S is at least 0.8Ω non domestic is a different kettle of fish. So where the incomer is say 0.25Ω should we allow for the supply to change due to DNO using different routes so a 32A supply instead of being 1.37Ω would need to be 1.27Ω to allow for the change in the incomer.

I think the 0.35Ω is (230 volt * 10% + 230 volt * 6%)/100 = 0.368Ω so it is the worst ELI line to neutral/earth that will be within the volt drop allowed.

 

[securespark]

When you say "at least", I was given to understand the Ze figures for TN systems were maxima.

But you can't plan your circuit for changes by the DNO on their side that are out of your control.

 

[JohnW2]

So we are also told TN-C-S domestic is at least 0.35Ω and TN-S is at least 0.8Ω non domestic is a different kettle of fish. So where the incomer is say 0.25Ω should we allow for the supply to change due to DNO using different routes so a 32A supply instead of being 1.37Ω would need to be 1.27Ω to allow for the change in the incomer.

As has been said, I presume you mean "no more than2, rather than "at least".

However, you raise an interesting question. In practice, of course, one accepts that a circuit is compliant if it has a low enough Zs as the time the Zs is measured, regardless of what changes in the supply might subsequently happen. That 'makes sense', but it is inconsistent with the way the regs appear to deal with breaking capacities of protective devices - unless it is a domestic environment with a 'type tested' CU, it is seeming required that the devices be capable of breaking the 'worst case' declared by the DNO (i.e. 16kA - i.e. Zdb about 0.014Ω), despite the fact that measured PFCs will often be dramatically below 16kA.

If their thinking about protective devices were consistent with that, they would presumably require that a circuit had to be designed to have an adequately low Zs with the 'worst case' Ze declared by the DNO - i.e. the 0.35Ω and 0.8Ω figures.

Kind Regards, John

 

[ban-all-sheds]

But you can't plan your circuit for changes by the DNO on their side that are out of your control.

411.4.5

 

[ericmark]

Yes less than not at least, my error, as to PSCC that is something else, I had to do something with a tower crane as it was too high, my thought was to use a fuse as the let through valve would be within limits before the RCBO however once fitted I found the resistance of the fuse was enough to bring it within limits so would not expect to find a problem when the DNO use a 100 amp fuse.

Again in the main earth problems are covered by the RCD however not all circuits are RCD protected, and so the question. When mothers house was moved from TT to TN-C-S it only just passed, but when rewired it passed with flying colours, so it would seem the ELI has changed.

Since we have micro generation setting need to allow for power to be exported so no longer can the DON use the full volt drop range, so unlikely as high as 0.35 any more, but can we use it?

 

[JohnW2]

Since we have micro generation setting need to allow for power to be exported so no longer can the DON use the full volt drop range, so unlikely as high as 0.35 any more, but can we use it?

I'm not at all sure that your reasoning as to how the 0.35Ω arises is correct but, even if it were, if the maximum Ze actually became less than 0.35Ω, the the answer to "can we use it (the 0.35Ω figure)?" would surely be 'yes', since to use an 'unnecessarily high' figure would actually be 'conservative', wouldn't it?

Kind Regards, John

 

[ericmark]

Sorry yes we can use the 0.35Ω if we add 23+13.8 then divide the product by the fuse size we have the maximum ELI for that fuse size with a TN-C-S supply, with other than TN-C-S the earth - line loop may not be the same as the neutral - line loop so only works with TN-C-S so 100A = 0.368 and 80A = 0.46 and 60A = 0.613333 so we should be able to rely on the ELI never being any higher.

However if the ELI is 0.25Ω when measured can we rely on it staying at 0.25Ω? That is the question. I seem to remember 0.80Ω was considered limit for TN-S and 18Ω for TT clearly plus the users earth rod. With over 18Ω you would need RCD protection and likely you would also need it with TN-S a cooker supply at 45A would need 0.97Ω so 0.17Ω for cable and MCB calculations show 31 meters of 10mm² but by time you add the volt drop across the MCB it would be less than that.

 

[JohnW2]

However if the ELI is 0.25Ω when measured can we rely on it staying at 0.25Ω? That is the question.

Indeed it is, but that is what we actually normally do - measure total Zs with a machine 'as is', at a particular point in time, and use that figure to determine whether the circuit is compliant as regards ADS, without any consdieration of what would happen if the Ze (below the 'declared maximum at the time of the Zs measurement) subsequently rose.

As I said, if the regs took the same approach as they apparently do in relation to 'breaking capacity', they would require us to design circuits so that they would still comply with ADS requirements if the Ze rose to the 'maximum declared value' - but they don't say that.

I seem to remember 0.80Ω was considered limit for TN-S and 18Ω for TT clearly plus the users earth rod.

That surely would make no sense, would it? Why should the component "other than the user's earth rod" suddenly go up from 0.8Ω to 18Ω - in fact, it should be less than 0.8Ω, shouldn't it, since only the DNO's L conductor would be part of the fault loop in the case of TT?

Kind Regards, John

 

[EFLImpudence]

I think the 0.35Ω is (230 volt * 10% + 230 volt * 6%)/100 = 0.368Ω so it is the worst ELI line to neutral/earth that will be within the volt drop allowed.

Am I missing something?

Apart from NOT being 0.35, isn't that approximation just a coincidence?

How can 23 Volts + 13.8 Volts have anything to do with the Ze, not to mention Zs with infinite variations of R1+R2?

 

[ericmark]

How can 23 Volts + 13.8 Volts have anything to do with the Ze?

The volt drop with the supply at 253 volts at the transformer must not allow the voltage to drop to below 216.2 volt at the DNO head, since with a TN-C-S supply the earth and neutral use the same conductor, then at 100 amp the loop impedance can't exceed 0.368Ω and the volt drop to stay within the volt drop limits. It is simple ohms law 36.8/100 = 0.368. In real terms unless only supplying a single premises then it would need to be below this figure, however how far below would require some complex calculations and so diversity so near impossible to say what the maximum figure really is.

However the way the DNO has set out the supply in many cases allows them to isolate sections by feeding from two or more points with fuse links used to select which is used, so the Ze is not a static figure, but can alter according to which fuses are in place. So recording Ze then recording Zs then retesting Ze will show if the system is within limits at the time of testing, but next day this could change.

Clearly with a Zs of 1Ω on a final ring we have no real worries, but what when we record 1.365625Ω right on the limit taking the new 95% into account. OK 1.4375Ω without the 95% but is that enough? With a 0.35Ω Ze then Zs likely needs to be better than 0.94Ω to stay within the volt drop on a ring final so with the maximum of 106 meters of 2.5mm² there should not be a problem. However how many times have you seen a reading far higher than that?

For a new installation no problem, 106 meters complied with so all A1, but what about when doing an EICR? at what point do you say, the ELI is too high? We all know if 1.37Ω it may pass, but the volt drop will not comply, however the 20 amp at centre and 12 amp even shared may be how we arrive at the design current for circuit Ib but there is no hard and fast rule about that. So if I say 13 amp at centre and 19 amp even on the circuit no one can say I am wrong or 32 amp at centre.

If you complete an EICR before a house is sold for the sellers, and some one else does an EICR after it is sold for the buyers, can any faults you failed to find be charged to you to correct? If you can show that the figures you recorded could very well be correct at the time of testing yet not correct on month after due to a permitted DNO change this is a get out of jail free card.

So although it may not actually change anything, understanding how readings can change is important.

 

[JohnW2]

The volt drop with the supply at 253 volts at the transformer must not allow the voltage to drop to below 216.2 volt at the DNO head .....

I doubt that they feel the need to, or are required to, consider the possibility that the total load on the transformer will (provided it is energised!) ever fall to zero (or anywhere near zero), so the voltage can still be above 253V at the transformer and yet not above 253V at the nearest consumer.

Kind Regards, John

 

[EFLImpudence]

No, I've figured it: (apologies if it was obvious)


23V + 13.8V is 16% of 230V

Max Zs is 230V / 5 [Type B multiple of In] x 80% (Ct) / In
which is 230V x 16% / In


Now as well we have a further Cmin of 95%.
which results in 34.96 / In

A coincidence - although 34.96 is not 0.35.

 

[John D v2.0]

There's double counting going on anyway, we have to take into account the whole loop impedance when checking, and check at 0.95 volts.
However the volts for the whole loop start at the transformer not the origin of the installation.
I suppose the added complication that not all the transformer load is flowing through the fault path makes it only "one point something" times counting not double. Also the impossibility of knowing the transformer voltage.

 

[SimonH2]

I doubt that they feel the need to, or are required to, consider the possibility that the total load on the transformer will (provided it is energised!) ever fall to zero ...

That would have been true perhaps a decade ago - it isn't now. On a bright sunny day* when everyone is out at work, it's possible for the power flows to reverse so the substation terminal voltage is now lower than at the end users' premises.
* It doesn't take many 4kW PV installations to "oversupply" a group of houses drawing maybe less than 100W each on average.

 

[JohnW2]

That would have been true perhaps a decade ago - it isn't now. On a bright sunny day* when everyone is out at work, it's possible for the power flows to reverse so the substation terminal voltage is now lower than at the end users' premises.

Maybe, although I would think that would be unusual.

Kind Regards, John