Upgrading supply/dno earthing

[JohnW2]

Good point, and easy to snipe from the sidelines. But , a minor works certificate lets you only record the Zs of the single circuit being worked on. The change of a consumer unit requires a full Electrical Installation Certificate, and this needs Ze to be measured and recorded.

Recorded - but not necessary 'measured' [ see Note (2) ] ...

This goes to show how important proper testing is, there may not be any Ze conductivity at all, because of the state is the service head. His EFLI measurements, and his house earthing, may be solely relying on bonding to gas and water services.

I've already acknowledged the latter, when I clarified that I had not meant to suggest that the service head should not be 'looked at' by DNO. However, as regards the former, as per the above, it is not necessary to measure Ze in order to completer a BS7671 eic.

Kind Regards, John

 

[Michaelsf90]

I love this place I know it needs a full electrical installation certificate. I know I need to borrow a test kit and measure the ze. But it seems for now more logical and easier to tell the dno I'm concerned about the earthing and its showing signs of corrosion and let them test it or change it or do something about it

 

[JohnW2]

I love this place I know it needs a full electrical installation certificate. I know I need to borrow a test kit and measure the ze.

You're right about the first thing but, as I have just illustrated, not about the second. You (or anyone elser) do/does not have to "borrow a test kit and measure the ze" in order to complete an EIC, since you are allowed to record on an EIC the Ze as 'determine by enquiry' (i.e. by asking the DNO - who will then give you a figure {without measuring anything} which will be the maximum permissible Ze for the type of earth - usually 0.35Ω for TN-C-S {"PME"} or 0.8Ω for TN-S).

But it seems for now more logical and easier to tell the dno I'm concerned about the earthing and its showing signs of corrosion and let them test it or change it or do something about it

Very much so. DNIOs are very used to people calling them because they have concerns (quite often unfounded) about aspects of the DNO equipment in their houses, and the DNO generally respond to such calls very quickly, without complaining (or charging!), even if nothing is wrong - they will usually rapidly attend your premises and 'inspect', but will not necessarily decide that they need to do anything.

Kind Regards, John

 

[RF Lighting]

Enquiring about Ze will get you the stock answer for that particular supply type and nothing else.

Only testing will show if the earth is actually functional. I have over the years worked on plenty of supplies where the suppliers earth has failed.

At the last testing and inspection training I attended, the preferred method of testing an installation is to dead test everything only, as it removes the risks associated with live testing. Everything apart from Ze which is still required to be done as a live test.

 

[JohnW2]

Enquiring about Ze will get you the stock answer for that particular supply type and nothing else.

Exactly. I can't disagree with that, since you appear to be repeating what I just wrote - which, as you saw, was ...

... you are allowed to record on an EIC the Ze as 'determine by enquiry' (i.e. by asking the DNO - who will then give you a figure {without measuring anything} which will be the maximum permissible Ze for the type of earth - usually 0.35Ω for TN-C-S {"PME"} or 0.8Ω for TN-S).

As you will realise, I was merely pointing out (for the sake of providing 'correct information'!) that TTC and Michaelsf90 were not right in saying/implying that one "had to" measure Ze in order to complete an EIC, since the BS7671 EIC form itserlf says that a 'by enquiry' figure is acceptable.

Only testing will show if the earth is actually functional.

Indeed so.

At the last testing and inspection training I attended, the preferred method of testing an installation is to dead test everything only, as it removes the risks associated with live testing. Everything apart from Ze which is still required to be done as a live test.

One can't disagree that such is the safest approach for the tester/inspector!

Does that mean that you never measure Zs on a circuit, instead recording 'calculated' figures on an EIC/EICR? If so, if one wanted to be pedantic, one might suggest that could presumably miss situations in which circuit Zs's were unacceptably high, despite a satisfactory Ze, because of problems (e.g. poor connections, or even absent connections!) in/around a CU etc.?

Were you taught to not even do live tests of polarity and phase rotation (where appropriate). To attempt to do the equivalent by 'dead testing' (presumably by 'tracing every conductor in the installation', by 'continuity' testing) would presumably be extremely tedious.

Do you even do Ze measurements on TN-C-S installations?

Kind Regards, John

 

[RF Lighting]

An R1+R2 low ohms dead test is used to calculate Zs, and will show any problems between the MET and the points of a final circuit. This also confirms polarity. You can still do a real Zs test at a socket outlet using a plug in lead if you wish.

Once a circuit or even a whole board is linked L>E the testing of each point takes the same amount of time as a live Zs test would. Ring circuits would be linked for the r1+r2 testing anyway, so it’s actually quicker to not live test rings.

As for phase rotation, honestly I can’t remember! But I guess the official line would be to prove this at the incommer as a live test at the same time as the Ze test, and then carry out dead continuity testing for each line to earth to confirm polarity and phase rotation at each point of use.

I high current test Ze on every installation. Whilst it’s rare, even TN-C-S supplies can sometimes be outside the spec due to a bad connection in the head or an excessively long supply run.

As we know, classroom teachings and real world situations may not always meet eye to eye.

 

[JohnW2]

An R1+R2 low ohms dead test is used to calculate Zs ...

Yes, that's obviously what I was referring to.

... and will show any problems between the MET and the points of a final circuit.

I'm not sure I understand that. Maybe you need to explain to me how you would normally measure R1+R2 on, say, a ring final circuit?

... You can still do a real Zs test at a socket outlet using a plug in lead if you wish.

I think I would personally want to, particularly if I had re-terminated any conductors after undertaking the R1+R2 measurement.

One interesting thing is that, at least if one follows the guidance of GN3, Ze is ('obviously') measured with any main bonding temporarily disconnected, whilst Zs is measured with such bonding still connected. In any installation that has extraneous-C-Ps, R1+R2+Ze is obviously going to differ, at least a bit, from (directly-measured) Zs - so which do you actually 'record' (as 'Zs'), if you do both?

That differing by "at least a bit" obviously becomes far more dramatic in an installation like mine. I wonder if you even bother to measure Zs (directly) in a TT installation? If you did in my house (with bonding connected), you would get figures typical of a TN installation (and, I think in all cases, low enough to be compliant in such an installation), although the Ze/Re you measured would usually be around 50Ω - 70Ω.

I high current test Ze on every installation. Whilst it’s rare, even TN-C-S supplies can sometimes be outside the spec due to a bad connection in the head or an excessively long supply run.

I suspected you would say that!

As we know, classroom teachings and real world situations may not always meet eye to eye.

Very much so. In so many fields, I have learned most of what I know about how things "are actually done" by watching, and talking to, relevant, 'experienced', professionals and tradesmen - and what I have learned from them has always been 'rather different' (in some cases, 'to say the least' ) from what I would have been taught in a classroom!

Kind Regards, John

 

[RF Lighting]

For a ring final circuit you can calculate your Zs from the highest r1+r2 reading recorded. This is a standard test which would be carried out regardless and is required as part of your initial verification.

On a radial circuit you would connect the line conductor of the circuit you are testing into the earth bar in the consumer unit and test line to earth at each point on the circuit. If the earth bar is not the MET then you’d also test between here and the MET to calculate Zdb.

I would always Ze a TT supply of course as they can and do fail.

I understand what you’re saying about Zs + Zdb will probably be lower than Zs + Ze, but it doesn’t actually matter. As long as your calculated worst case scenario values meet the minimum required value then you have proved the installation to be safe.

In my opinion this is the proper way to prove the safety of your installation as the parallel paths of your bonding can not be relied upon - they could be removed by a neighbour tomorrow without your knowledge for example.

 

[JohnW2]

For a ring final circuit you can calculate your Zs from the highest r1+r2 reading recorded.

Yes, I obviously understand that, but I was asking HOW you undertake the R1+R2 measurement. In particular, if it involves removing (and subsequently replacing) the CPCs from the earth bar, then, unless you subsequently undertake a ('live') Zs measurement, you theoretically can't be certain that the Zs has 'survived your fiddling in the CU'.

I would always Ze a TT supply of course as they can and do fail.

Yes, of course. However, I was asking whether you would normally 'bother' to undertake ('live') Zs measurements in a TT installation - since not only would it probably be 'unhelpful' (since, as we have both said, one cannot rely on incidental paths to earth remaining present) but, in a house like mine, it could yield some 'surprising' results.

I understand what you’re saying about Zs + Zdb will probably be lower than Zs + Ze, but it doesn’t actually matter. As long as your calculated worst case scenario values meet the minimum required value then you have proved the installation to be safe.

Yes, I agree with all that. However, I was really asking about the 'mechanics' of completing an EIC/EICR. Are you saying that, since it represents the 'worst case scenario' (i.e. the situation which would exist if all the bonded extraneous-c-ps were to disappear), you would record the 'calculated' Zs figures, even if you had measured Zs (and got somewhat lower answers).

In my opinion this is the proper way to prove the safety of your installation as the parallel paths of your bonding can not be relied upon - they could be removed by a neighbour tomorrow without your knowledge for example.

Exactly - which is why I asked whether you bother to measure Zs in a TT installation. One would hope that any TT installation is (like mine) designed not to rely on OPDs for fault protection (i.e. such protection is provided by RCDs), even if, 'at present', the loop impedances are such that it could.

However, one other thought about relying on 'calculated' loop impedances (i.e. from R1+R2 and Ze). Measurement of 'very small' resistances (like R1+R2) is a very iffy/unreliable business, and I imagine that Zs measurements (when it is current that is being measured) are probably much more reliable. For that reason, I suppose that, were it not for the potential 'safety' issues, the ideal would be to undertake ('live') Zs measurements with the bonding disconnected, rather than relying of R1+R2 measurements.

Kind Regards, John

 

[RF Lighting]

If I was testing a full board then I’d simply link the busbar to the earth bar using a lead with crocodile clips on and start testing.

If you’re testing a single circuit then you can either connect the line into the earth bar or fit a temporary test link. No need to disturb the earth bar once you’ve done your R1 + R2 test, or just R2 test every point if you are going to prove polarity a different way.

There’s no point taking a live Zs reading. I’ve already proved that every circuit has a good earth and is within the design specs.

On a modern installation every circuit will have RCD protection so you can only carry out low current loop testing anyway so there’s nothing to be gained.

You can record either measured or calculated figures on the certificate.

What is iffy about low ohms testing? I have a calibrated meter which gives consistant results, and can be verified by calculating the estimated length of the cable run multiplied by the resistance per meter values of said cable.

 

[EFLImpudence]

Is adding two (or more) measurements a calculation or a measurement?

 

[RF Lighting]

It’s a calculation.

 

[EFLImpudence]

Are you sure?

If I only have a 5m. tape and want to find the length of some thing over 5m. and I measure it as 5m. plus 4m., is the 9m. merely a calculation?`

 

[JohnW2]

Is adding two (or more) measurements a calculation or a measurement?

It’s a calculation.

Are you sure? If I only have a 5m. tape and want to find the length of some thing over 5m. and I measure it as 5m. plus 4m., is the 9m. merely a calculation?`

Oh dear. It's not even the weekend, but this seems to be incredibly semantic/pedantic, particularly given that we all know what distinction we are talking about!

As per EFLI's example, if one were talking about separately measuring R1 and R2 and adding them to get R1+R2, then I think one could reasonably call that a 'measurement' (even though, in some such situations, the accuracy might differ according to how many 'actual measurements' had been undertaken).

However, in context we are talking about, "by calculation" simply as shorthand for describing the process of measuring (R1+R2) and Ze separately (by two totally different methods) and adding them together to get the loop impedance. The "by calculation" shorthand does not detract from the fact that the things being added (in 'the calculation') are both 'measurements'.

If my car travels at a constant speed for a period of time, there are two ways in which I can determine the speed - I can either look at the speedometer (which represents a direct measurement of speed) or else I can measure the distance travelled and time taken and then divide one by the other to get speed. Whilst the latter still just involves two 'measurements', I don't think that there would be anything particularly wrong than saying that the determination of speed had been performed 'by calculation' (a calculation performed upon two measurements).

... but now, hopefully, back to the actual discussion!

Kind Regards, John

 

[EFLImpudence]

It is relevant as you were discussing measuring and calculating Zs.

I would say a calculation would be knowing the length of a cable and the resistance per meter and working it out; anything done using meters is a measurement.