Enabling EPS DIY?

[Oldgreybeard]

BS 7430 advises 20 ohms for generating sets.

The interesting thing here is that 20 ohms is far too high for just overcurrent protection to be safe, and unnecessarily low if there is 30mA RCD protection. It seems to be a "neither fish nor fowl" number, plucked from thin air (and not unique in that regard).

The Ze limits for non-RCD protected installations (0.35 ohms for TN-C-S/PME, 0.8 ohms for TN-S) are set to allow over-current protection to also provide protection from electric shock in the event of a fault. The typical PME Ze of around 0.3 ohms ensures that even a 32A circuit breaker will trip before the MET voltage rises to an unsafe level (the MET touch voltage would only be 9.6V with a 32A earth fault current and a Ze of 0.3 ohms (ignoring installation conductor resistances, R1 and R2 just to make the point).

A 20 ohm Zs results in the MET hitting the 50V maximum touch voltage (which is stipulated in BS7671) at a fault current of just 2.5A. This means that for any generator rated at more than about 600VA, that doesn't have RCD protection, there is a risk that the MET touch voltage could exceed the 50V limit if there were a fault.

[Swwils]

All island mode systems are TN-S.

You just also remember the size of the system leakage capacitance is negatively affected by the length of the wire and the number of loads or the resulting impedance between the active conductor and earth (Z becomes smaller). For this reason, the instruction to limit this protective measure to the supply of an electrical consumable applies generally, and the length of the wire must also be borne in mind. For fault protection, additional measures are still needed ontop of this.

[Oldgreybeard]

All island mode systems are TN-S.

You just also remember the size of the system leakage capacitance is negatively affected by the length of the wire and the number of loads or the resulting impedance between the active conductor and earth (Z becomes smaller). For this reason, the instruction to limit this protective measure to the supply of an electrical consumable applies generally, and the length of the wire must also be borne in mind. For fault protection, additional measures are still needed ontop of this.

Hence part of the reason we always measure R1 and R2 when testing any installation (main part of R1 and R2 testing is integrity checking, though). Nothing new, it's been a mandatory requirement to measure R1 and R2 for as long as I can remember. It was certainly taught when I did the C&G testing course years ago (what is now 2391).

[AGT]

2.4 ohms max earth rod resistance for EV charging at 32 amps.
Can see many people achieving that

[Oldgreybeard]

2.4 ohms max earth rod resistance for EV charging at 32 amps.
Can see many people achieving that

Indeed, can't be achieved with a normal rod, they just don't have the surface area to get close to that figure. They use copper earth mats at transformers to get the resistance to a really low value.

Given that the maximum earth fault current is only 30mA though (because of the RCD protection that's required) it's pretty much a moot point, though. As long as the rod and associated wiring in the fault loop can conduct 30mA to earth without the potential rising above 50V on anything that can be touched it's safe enough in practice.

[Kenny000666]

Did a test with the battery connected (arrived today) and backup enabled.

When the AC goes off, the voltage drops from 249(ish)V to 231v. But at no time does it go to 0 volts (well according to my multimeter and my watching it )

When the AC comes back it after 20seconds or so, the voltage goes back to 249V.

Looks like it’s instantaneous. Next test connect a test pi into the socket and see if it reboots between losing power.