Ze of 0.28ohms on L1, 0.16ohms on L2 & 0.15ohms on L3

Sounds like a bad connection on L1 to me. If it's PME/TN-C-S drawing 200A from L1 will give a voltage drop of around 56V - which is well beyond normal limits.

  - Andy.

AJJewsbury said:

which is well beyond normal limits

Certainly outside ESQCR would normally be 36.8 V maximum volt-drop (for the range 216.2 to 253 V to Earth. ... although with three-phase I guess they'd look at the line-line voltage?

What is the maximum demand?

Hi gents,

thanks for the reply, I believe it is an issue with the phase in the Transformer, but again the DNO will not look at it which I think is obscene. 

The maximum demand for the project is around 140A per phase plus spare capacity. 

well the normal safe thing to do for cables that need fast fault protection when Zs is high is an earth fault relay or a breaker with built in earth fault detection, or if it had been a bit lower current then RCDs.

But in this case load it up with a few kW and see how much the line really droops. At 140A, and 30V drop, it 4000W goes somewhere so it will either warm up and weld itself good, or warm up and blow open. In the first case good, in the 2nd case, the DNO then have to come out, then  followed by 'good'' again.

In the past I have been fooled by faults measured at low current test gear that self heal once a real load arrives.

Mike.

but again the DNO will not look at it which I think is obscene.

If it is a TN-C-S supply then describe the problem to the DNO in terms of voltage drop rather than Earth loop impedance - they're legally obliged to stay within ESQCR limits.

You could also point out that 0.28Ω is outside of most DNO's limit for EFLI for 200A (UKPN say 0.2Ω and WPD 0.22Ω for instance - there's probably a common Gxx document somewhere).

   - Andy.

https://www.nationalgrid.co.uk/downloads-view-reciteme/607407

National Grid earth loop impedance testing of services - acceptable values in section 3.0

This standard assumes most of the resistance is in the non current-carrying earth part of the loop. In this case a good tenth of an ohm is in at least one of the phase paths - and once loaded to 100A plus if it is a a problem of a single corroded point contact , something will be glowing in the dark. Power in 0.1 ohm at 100A is 1kW- that's like a small welder. !!

Again, I'd apply test loads and see if the LN voltage drop you would expect from that resistance is really observed (you may see something on the meter even with a kettle and or a few kW of heaters), if it is, that is a much bigger point to raise to justify remedial works than just the earth impedance  blowing a fuse in 8 or 9 seconds instead of 4 or 5 (regs fail but not an immediate danger).

Realise that the DNO will be approaching this in the same way Thames Water approach maintenance - they have miles of cables & thousands of transformers to look after, and they do the worst ones first. Also realise that any management that did not prioitize the best interests of the shareholders would be fired at the end of the financial year, so generally there is little appetite for doing upgrade work just in case, it has to wait until it is really needed.

Mike.

JL97 said:

The maximum demand for the project is around 140A per phase plus spare capacity. 

OK, so is the issue that there's a distribution circuit of quite some length between the 200 A fuses and the first switchboard? Because for total load of 140 A, final circuit protective devices will be OK with that ... or is it a design condition for the switchboard?

I agree it looks like there's a faulty joint somewhere, but how much time and resources will they allocate to tracing it?

I guess it all depends on the man on site - if they want to wriggle out of doing anything until there's a "real" problem they may on the basis that it's within limits, albeit significantly different from the other phases.

OK, so is the issue that there's a distribution circuit of quite some length between the 200 A fuses and the first switchboard? Because for total load of 140 A, final circuit protective devices ought to be OK with the Ze we are talking about ... or is it a design condition for the switchboard?

The latest version of P23 would seem to support the DNO's position, that the Ze is acceptable for the service rating.

As others have said, though, that doesn't mean the voltage drop meets ESQCR.


SOLUTION ... unless the switchboard manufacturer requires anything different ...  is no different to that used in a 100 A single-phase domestic installation, where often the same conditions may apply (unable to operate the 100 A service fuse in 5 s where Ze > 0.38 ohms, e.g. TN-S, or 1 s for TT), i.e. upto the switchboard, all you need is a wiring system that is equivalent to  double or reinforced insulation, and you are then ADS is not necessary, only protection against thermal effects (overcurrent and fault current protection via adiabatic). This is acceptable for protection against electric shock, because you are using Section 412, rather than Section 411.


I'm guessing, though, that a design has been undertaken assuming something different?