Earth Loop Impedance

[jg321]

also the supply can be repaired and it seems to say they can turn a TN-S into a TN-C-S without informing the user

Interestingly, when an isolator was being fitted to my supply a while back, I got the chance to see inside my cutout, and despite it looking a lot like TN-C-S from the outside, it is in fact TN-S as the feed is split con with no link between N&E. Interestingly I appear to be daisy chained to next door (as the supply cable goes off under my floor towards the neighbours, and through the party wall into their property), so who knows what their cutout looks like in terms of earthing arrangement.

When I asked the guy fitting the isolator, he said that although it technically is TN-S at my premises, they're told to treat everything as PME anyway, as it either has, or will be converted to PME at some point anyway. Not sure how much truth there is to this, and I don't have the kit to measure Ze, although I do have some paperwork (MWC) which came with the house, stating the Zs on the downstairs ring final was measured at 0.8.

 

[JohnW2]

... Not sure how much truth there is to this, and I don't have the kit to measure Ze, although I do have some paperwork (MWC) which came with the house, stating the Zs on the downstairs ring final was measured at 0.8.

If that Zs was measured (as it normally would be) at the mid-point of the ring, then one imagines that the Ze would be pretty low, almost certainly below the 0.35Ω 'maximum' quoted in relation to TN-C-S.

Kind Regards, John

 

[John D v2.0]

That's what I wrote, isn't it? I wrote "... very fast rate of rise ... with high frequency components", not "infinitely fast rate of rise ... with infinitely high frequency components"

Yes i just wanted to emphasise that, as i suspected that was the difference in our views.
This is even clearer based on the following quote:

It's interesting that, for such a short disconnection time, the situation would actually be 'worse', by a factor of √2 than would be the case if one undertook ('naive'?) 'RMS calculations'.

It would, but only on the assumption that the fault current rose instantly in proportion to the voltage

 

[JohnW2]

It would, but only on the assumption that the fault current rose instantly in proportion to the voltage

How quickly the fault current rose (following 'instantaneous' appearance of a fault) would, as you have said, depend upon the inductance of the fault circuit. That is generally (alwaysusually always) not going to be known, so I don't see how it could be taken into account in any 'generic' calculation.

Kind Regards, John
Edit: silly typo corrected!

 

[John D v2.0]

That is generally (always always) not going to be known, so I don't see how it could be taken into account in any 'generic' calculation

As you say we are just worried about worst cases. I'm sure the real answer is different, but the standard lv transformer has probably been tested to work out the worst case. The additional inductance in a real circuit is probably more, but the savings in copper to make the cpc even smaller in t&e are probably not worth it.

 

[JohnW2]

As you say we are just worried about worst cases. I'm sure the real answer is different, but the standard lv transformer has probably been tested to work out the worst case. The additional inductance in a real circuit is probably more, but the savings in copper to make the cpc even smaller in t&e are probably not worth it.

[apologies for silly typo on previous post - now corrected!]

I remain pretty confused about this 'let through' business, not the least by the 'let through charts' to which you have referred. I presume you are talking about something like this (Wylex) ...

For a start, I still don't see why there is a need for these charts to be in an 'unfamiliar' format. Since we've been talking about a B6, I've done my best to read figures off the 6A curve in the above and, once I had done that, it is a trivial exercise to turn it into the 'familiar' "t vs. I" characteristic curve format, namely ...

You will see that, even at 800 A, t is only about 1.6 ms, and when the current gets up to 10,000 A it is down to about 0.16 ms (i.e. 160 μs). I have to believe it but, in view of my previous comments about 'mechanical factors' etc., I'm pretty surprised to see that an MCB can disconnect that quickly.

However, the main thing is that I cannot find any evidence that the PSCC is anything other than what we would expect it to be, nor that that chart is presenting anything other than the characteristics of the device - so, with reference to you comment above, not related to any particular LV transformer (or anything else external to the device).

On the other hand, if, despite the absence of any mention of it, the 'PSCC' in Wylex's 'chart' is meant to be something other than Uo divided by the ('worst case') relevant fault loop impedance, we're back to square one, even with the 'chart', since we don't know how to determine that 'PSCC'.

Kind Regards, John

 

[ericmark]

As you say confusing, in the main MCB's and RCBO's are either 4.7 kA or 10 kA rating, all we want to know is if a 100A BS88 fuse is fitted will the let through amps exceed 4.7 kA or not so as the installer I know if I can use 4.7 kA rated devices or not. I really don't want to have to read curves.

As said only once have I measured the PFC of an installation and found it exceeded 4.7 kA, I mounted a BS 88 fuse holder and fuse inside the enclosure, but could not really work out if it did or did not now comply, however the resistance of the fuse resulted in the PFC to now be under 4.7 kA so no worries.

With a 100A BS88 it says watts loss 10W so 0.001Ω, need 0.05Ω to be under 4.7 kA at 230 volt, however how accurate is a loop impedance meter when measuring 0.05Ω? For 10 kA 0.023Ω but meter will measure 0.02Ω, I have never measured a loop impedance anywhere near that low on a domestic. For a 500 kVA transformer looking at 0.0166Ω so plus BS88 fuse and some cable it is unlikely to be lower than 0.02Ω.

 

[JohnW2]

As you say confusing, in the main MCB's and RCBO's are either 4.7 kA or 10 kA rating ...

Where does 4.7 kA come from? I think all the 'standard' domestic MCBs/RCBOs I've seen have had a breaking capacity of 6 kA.

... all we want to know is if a 100A BS88 fuse is fitted will the let through amps exceed 4.7 kA or not so as the installer I know if I can use 4.7 kA rated devices or not. I really don't want to have to read curves.

Yes, that what (and all) we need to know. The concept of 'let through amps' doesn't really exist - the only way a fuse can influence the fault current is by the resistance/impedance it adds to the circuit, and if that is sufficient that the PFC falls to below the breaking capacity of the MCB, then that is obviously fine.

My understanding is that the 'dispensation' to which you refer (when theoretical PFC exceeds the breaking capacity of an MCB, but there is a BS88 upstream of the MCB - as per the 'dispensation' for domestic CUs) is more complicated than you suggest, and a while ago we considered a theoretical article which went into this in some detail.

Kind Regards, John