Volt drop, loop impedance and micro generation?

[ericmark]

I am trying to get my head around the new loop impedance figures. We would set the step down transformer to an open circuit voltage of 230 volt plus 10% = 253 volt and we were allowed a volt drop of 230 minus 6% = 216.2 volt but the loop impedance at this point can not exceed 0.35 Ω.

A B type MCB needs 5 times the rated current to flow so taking a B32 as an example at 253 volts a loop impedance of 1.58125 Ω is required. Since we measure total earth loop impedance not just that within the installation it is the voltage at the transformer which matters not the voltage at the house.

The volts could be 230 which would mean the loop impedance could be 1.4375 Ω however at the transformer it is unlikely the volts would have dropped below that level. Until we start to consider micro generation then in order to feed into the grid the transformer voltage may be set lower.

Clearly at night solar panels will not generate and wind power will also at some point stop even the sterling generators fitted to central heating boilers will likely not be running in the summer so the transformer will never be set to 216.2 volts although the supply to house may reach that point.

So case scenario is 216.2 volts with a loop impedance of 0.35 Ω (TN-C-S) so with a 100A supply the transformer would need to be 35 volt above that value so transformer volts would be 251.2 volts. For the transformer volts to be set at 230 the loop impedance would need to be 0.15 Ω approx.

I realise micro generation can mean the step down transformer is set to a lower voltage and that we are not considering just one house but a line of houses but I fail to see how the short circuit current at a loop impedance of 1.44 Ω would drop below the 160A mark even with micro generation lowering the transformer voltage.

I realise it requires Norton's theorem or Thévenin's theorem to truly work out the short circuit current it is not a single house. But just can't see why we have the reduction?

 

[conny]

Don't understand the question but the answer is 42.

The answer is always 42

 

[JohnW2]

I am trying to get my head around the new loop impedance figures. .... A B type MCB needs 5 times the rated current to flow so taking a B32 as an example at 253 volts a loop impedance of 1.58125 Ω is required. Since we measure total earth loop impedance not just that within the installation it is the voltage at the transformer which matters not the voltage at the house.

That's an important point which seems to often get overlooked (probably for practical reasons, since one never knows the voltage at the transformer).

I can but presume that the new figures take into account the (very unlikely, but theoretically not impossible) situation in which the voltage at the transformer was only U0-5% (i.e. 218.5V).

Kind Regards, John

 

[EFLImpudence]

with a loop impedance of 0.35 Ω (TN-C-S)

Unless it isn't.

 

[ericmark]

I am trying to get my head around the new loop impedance figures. .... A B type MCB needs 5 times the rated current to flow so taking a B32 as an example at 253 volts a loop impedance of 1.58125 Ω is required. Since we measure total earth loop impedance not just that within the installation it is the voltage at the transformer which matters not the voltage at the house.

That's an important point which seems to often get overlooked (probably for practical reasons, since one never knows the voltage at the transformer).

I can but presume that the new figures take into account the (very unlikely, but theoretically not impossible) situation in which the voltage at the transformer was only U0-5% (i.e. 218.5V).

Kind Regards, John

It would seem you have seen my point. I considered working it out but Norton's theorem and Thévenin's theorem are not my strong points.

 

[JohnW2]

I am trying to get my head around the new loop impedance figures. .... A B type MCB needs 5 times the rated current to flow so taking a B32 as an example at 253 volts a loop impedance of 1.58125 Ω is required. Since we measure total earth loop impedance not just that within the installation it is the voltage at the transformer which matters not the voltage at the house.

That's an important point which seems to often get overlooked (probably for practical reasons, since one never knows the voltage at the transformer). ... I can but presume that the new figures take into account the (very unlikely, but theoretically not impossible) situation in which the voltage at the transformer was only U0-5% (i.e. 218.5V).

It would seem you have seen my point. I considered working it out but Norton's theorem and Thévenin's theorem are not my strong points.

Nor mine - but, frankly, with the amount of information we usually have, I'm far from convinced that use of those theorems would necessarily help.

On the other hand (although I haven't fully thought this through), by undertaking additional measurements (the likes of which most of us are not directly equipped to do) it would probably be possible to get a handle on the voltage at the transformer, hence the true PFC and PSCC.

Kind Regards, John

 

[bernardgreen]

My opinion is that these "absolute" figures of impedance are far from absolute since the circumstances will vary from place to place and depending on the amount of electrical power being supplied by that branch of the network..

The regulations for house wiring take account of the temperature of the cable, cable bunching and hoe the resistance of copper changes with temperature.

Likewise the resistance of the cables under the road will change will temperature so the loop impedance is going to change. And some of the cables under the street do get quite warm at times of peak load.

And the regulations seem to consider a short circuit fault will be a zero ohm fault. ( do they somewhere specify the maximum and minimum resistance ( impedance ) for a BS 7671 short circuit.

While I totally agree that a dead short zero ohm (*) fault must be able to pull enough current to trip the breaker within one half cycle I do wonder about way the regulations handle the matter. Especially as most accidental short circuits will blow their zero ohm path and become very low impedance faults that may not be pulling 5 times the breaker's rated trip current. Hence the breaker will operate on its thermal trip.

(*) zero being a few micro ohms unless the fault is super conducting at zero Kelvin ( minus 273 degrees C )

 

[ericmark]

I do agree there will be variations in the loop impedance and as a result some lee way is required.

But in the main the earth loop is not that important as the RCD will disconnect with a direct short to earth however with a passive type there needs to be enough voltage to operate the RCD an with a TN system a direct short to earth could in theroy mean there is not enough voltage to operate the RCD.

With this in mind the loop impedance at the consumer unit is very important. I am sure there should be for every RCD a loop impedance figure which when supplied through a 100A fuse will ensure either the fuse will rupture or the RCD will operate but I have not seen such figures published.

With a line - neutral fault the RCD offers no protection but I would hope where the magnetic part fails to operate the thermal part will disconnect before any damage to cable.

At the socket we are well aware of the volt drop problems and the possibility the RCD will not work as a result hence why we use active types.

But at the consumer unit we can have zero volts and the RCD will remain on. I know there was a problem pre-electronic type 30 mA would trip the RCD at 230 volt but even at 50 volt the RCD would still work although it may have required 5 times 30 mA for it to work.

Once we reach 50 volt to protect against shock we don't really need the RCD to work and so we rely on the MCB to disconnect to pull the voltage down that much there would need to be a huge current flow so with an earth loop impedance of 0.92 even a B50 MCB will trip. With the loop impedance at the consumer unit at 0.35 that would mean 142 volts at the consumer unit which should work any electronic RCD. As we look at type C or D likely the supply fuse will rupture first.

So it would seem the RCD will protect. I often feel the don't rely on RCD is the wrong way around. RCD's are tested there is no way to test a MCB so may be we should not rely on MCB's?

 

[bernardgreen]

with a passive type there needs to be enough voltage to operate the RCD an with a TN system a direct short to earth could in theroy mean there is not enough voltage to operate the RCD.

Do the RCD manufacturers specify the minimum supply voltage necessary for the electronics in the RCD to have enough power to operate.


Hopefully the electronics will operate correctly if there is only 50 volts between Live and Neutral. If the RCD power supply to the electronics requires more than 50 volts then the device is unlikely to operate when a hazardous voltage exists ( greater than 50 volts is hazardous ) and an earth leakage ( possible via a person's body ) is occurring. At least the majority of RCBO manufacturers have taken account of the fact that losing the Neutral renders an electronic device inoperative and have therefore incorporated a "functional" earth as an alternative source of "Neutral". If there is a loss of Live doesn't then it does matter if the RCBO doesn't operate as without Live there is no hazardous voltage on the circuit.

A non electronic RCD requires no power supply to operate the trip, the energy required to operate the trip comes from the secondary winding of the sensor. Hence these will work what ever the supply voltage.

 

[JohnW2]

My opinion is that these "absolute" figures of impedance are far from absolute since the circumstances will vary from place to place and depending on the amount of electrical power being supplied by that branch of the network.

Sure, the impedance will be temperature-dependent, hence load-dependent, and therefore will vary. The 'worst case' (in terms of fault protection) will exist when the local part of the network is 'maximally loaded' (hence hottest). I think that the DNOs say that the 'maximum' Ze is 0.35Ω for TN-C-S and 0.8Ω for TN-S. If that is the case, then I wonder if those 'maximum' figures do, indeed, relate to 'maximum loading' of the the local network?

And the regulations seem to consider a short circuit fault will be a zero ohm fault. ( do they somewhere specify the maximum and minimum resistance ( impedance ) for a BS 7671 short circuit.

They talk in terms of a fault of "negligible impedance" - which, AFAIAA, they do not define.

While I totally agree that a dead short zero ohm (*) fault must be able to pull enough current to trip the breaker within one half cycle I do wonder about way the regulations handle the matter. Especially as most accidental short circuits will blow their zero ohm path and become very low impedance faults that may not be pulling 5 times the breaker's rated trip current. Hence the breaker will operate on its thermal trip.

I'm not sure that the faults would necessarily "blow their 'zero' ohm path" before the OPD has operated, would they? However, even if they did, I'm really not sure what one can do about it, in terms of OPD protection - faults of 'a handful of ohms' are always going to be a problem. In the case of L-E faults, this is perhaps a strong reason for having RCD 'back-up'.

Kind Regards, John

 

[JohnW2]

So it would seem the RCD will protect. I often feel the don't rely on RCD is the wrong way around. RCD's are tested there is no way to test a MCB so may be we should not rely on MCB's?

I have a fair bit of sympathy with that view - and it could well be that this will eventually change. I suppose one of the issues is that one also needs protection against L-N faults and, given that R1+R2 within an installation will invariably be greater than R1+Rn, so that (in a TN installation) a circuit with satisfactory protection against L-N faults will also usually have satisfactory protection against L-E ones - so therereally should never be a 'need' to rely on RCDs.

Of course, as we gradually change to RCBOs, unless they give some indication about 'the mode of the trip', one will not know which part of it has tripped! That might present some problems of fault finding/tracing!

Kind Regards, John

 

[JohnW2]

Do the RCD manufacturers specify the minimum supply voltage necessary for the electronics in the RCD to have enough power to operate.

I've never noticed them specifying that but, as you go on to say ....

Hopefully the electronics will operate correctly if there is only 50 volts between Live and Neutral.

... and when one looks at the way that devices derive the power for their electronics, one would expect that it would still work down to quite low voltages.

At least the majority of RCBO manufacturers have taken account of the fact that losing the Neutral renders an electronic device inoperative and have therefore incorporated a "functional" earth as an alternative source of "Neutral".

True, but that is addressing the seemingly incredibly improbable scenario of an L-E fault (or L-E current though a human being) arising at the very moment that the neutral was 'lost' (itself an extremely rare occurrence). There is no inherent difference in this respect between an (electronic) RCBO and an (electronic) RCD, but I have personally yet to see any RCD taking the same 'precaution'.

A non electronic RCD requires no power supply to operate the trip, the energy required to operate the trip comes from the secondary winding of the sensor. Hence these will work what ever the supply voltage.

True, but I wonder about your tense - I personally haven't seen a non-electronic 'domestic' RCD for many years, have you? As I recently wrote, I suspect that one of the problems of non-electronic ones was a lack of precision/consistency as regards IΔn.

Kind Regards, John

 

[bernardgreen]

I am aware of non electronic "RCD" being used on supplies to equipment where any difference in current between Live and Neutral must trip the supply. The difference could be caused by leakage to earth or ( more important ) from another source of voltage when there is no voltage on the supply to the RCD.

It wasn't a case of ""must trip at 30mA and must not trip at 29mA"" but at any detectable difference between Live and Neutral.

To be fair it is not the only "leakage" protection on these supplies but is considered to be the most reliable "leakage" detection.

 

[JohnW2]

I am aware of non electronic "RCD" being used on supplies to equipment where any difference in current between Live and Neutral must trip the supply. The difference could be caused by leakage to earth or ( more important ) from another source of voltage when there is no voltage on the supply to the RCD. ... It wasn't a case of ""must trip at 30mA and must not trip at 29mA"" but at any detectable difference between Live and Neutral.

That's surely unrealistic, and imprecise. There simply has to be a level of L-N current difference below which it couldn't trip - maybe not 30mA, 10mA or even 1mA, but some finite value.

Furthermore, in context, if one did have a requirement for a device which would trip with extremely small L-N current differences, then surely one would need an 'electronic' one, wouldn't one? With a non-electronic RCD, the energy to operate the mechanical trip mechanism derives entirely from the 'imbalance current' in the secondary of the transformer. If that current were incredibly small, it surely would not be able to trip the mechanism, would it? ... whereas an 'electronic' one could.

Kind Regards, John

 

[bernardgreen]

If that current were incredibly small, it surely would not be able to trip the mechanism, would it?

The sensor has several turns of the Live and Neutral around the toroid and does give enough energy on a low differential current.

... whereas an 'electronic' one could.

only if there was a power supply to the electronics.

The requirement was to disconnect the incomer even if there was no voltage on the incomer. Leakage being due to voltages generated on the site.