AC EVSE (without vehicle) has the following touch-current limits according to BS EN IEC 61851-1:2019 and BS EN 61851-22:2002.

The test is made when the EVSE is functioning with a resistive load at rated output power.

In addition, BS EN 61851-21 permitted up to 5 mA of additional "leakage" from the vehicle.

Therefore, up to 8.5 mA AC may be expected - I always tell people to expect 9-10 mA combined because occasionally, it's on the verge of "might just operate" the 30 mA RCD mandated by BS 7671. There's certainly no overhead for other equipment on the same circuit.

Plus on the protective conductor itself (between PC and PE) there is the DC/pulsed DC of the pilot, don't forget.

This is a new one to me, and it seems quite oddly worded - as I presume you have a copy, could you clarify what is a 'network pole' in this context please ?  It also reads as if touch current is interpreted quite differently to the way we would for an appliance and really means non diverted CPC current.

M

mapj1 said:

could you clarify what is a 'network pole' in this context please

Live conductor - not uncommon.

mapj1 said:

It also reads as if touch current is interpreted quite differently to the way we would for an appliance and really means non diverted CPC current.

I think you can view it like that, but again this is becoming more common in product standards.

Sorry, I am not quite picking up on what Mike is saying, can you just clarify if this is touch current as measured through a body model?

I think, which is why I queried it, they seem to allow any no of amps to flow in the CPC, so long as it is not flowing L-E, In other words L and N currents balance - i.e. in a standard worded like that,  you are not regulating  the shock a victim would receive if the CPC broke and their body replaced it, but the rather the result of an RCD-like measurement of residual supply current, phase currents minus neutral currents  sort of thing. 

On a stand alone item like a kettle on an insulated worktop, the two are the same, but not for anything that has any degree of installation that allows a second earth path.

As a silly extreme imagine a double wound transformer with its primary wired L-N and its secondary wired between the supply CPC and some plumbing or building steel that is casually earthed , The CPC current is not derived from a pole of the supply, so L-N currents balance but a CPC current  is very much flowing , and if the voltages and series resistances were chosen right then breaking the cpc and holding both sides could be painful ;-) 

As far as I know most PA testers actually do measure CPC current, with varying degrees of error on non-sinusoidal waveforms, and for the purposes of the CE mark, the test gear that gets used in the labs does too - well at least it does on the stuff we CE mark,which is not electric  car chargers to be fair.

It just seems a strange way to set a limit that is really about shock protection, hence the query.

Mike.

mapj1 said:

I think, which is why I queried it, they seem to allow any no of amps to flow in the CPC, so long as it is not flowing L-E, In other words L and N currents balance - i.e. in a standard worded like that,  you are not regulating  the shock a victim would receive if the CPC broke and their body replaced it, but the rather the result of an RCD-like measurement of residual supply current, phase currents minus neutral currents  sort of thing. 

Agreed ... that is effectively the approach taken in some standards.

mapj1 said:

As a silly extreme imagine a double wound transformer with its primary wired L-N and its secondary wired between the supply CPC and some plumbing or building steel that is casually earthed , The CPC current is not derived from a pole of the supply, so L-N currents balance but a CPC current  is very much flowing , and if the voltages and series resistances were chosen right then breaking the cpc and holding both sides could be painful ;-) 

Yes, I do see where you're going. But that's a number of things going wrong in sequence rather than instantaneously, so therefore not "single fault conditions" or "additional protection".

mapj1 said:

It just seems a strange way to set a limit that is really about shock protection, hence the query.

BS EN IEC 62368-1 is very interesting. Doesn't align with BS EN 61140 at all really - especially where "SELV" and "PELV" are concerned, hence limiting the use of electronic supplies for certain special locations in BS 7671:2018+A2:2022.

Is it fair to say it appears that the protective conductor currents may approach the threshold specified in Regulation 543.7, which pertains to high protective conductor currents exceeding 10 mA ? 

yes but actually it sounds like it could be a lot higher, the protective conductor currents may be anything - amps if you like!

As apparently only the CPC current that comes L-E directly is being considered. other currents flowing in the CPC that may still be a consequence of the operation of the charger or the car attached, between the CPC and other paths to earth such as mechanical mountings or external reference electrodes etc are ignored.

I can see an argument for ignoring diverted neutral current coming from the network, as there is very little one can do about that as product design, apart from TT,  but it feels odd to ignore currents in the CPC generated by the system under test, especially in a system that uses the bit of CPC between car and charger  as part of its signalling path.

Mike.

“currents in the CPC generated by the system under test” Is this matter typically disregarded, or is it simply not given due consideration?

mapj1 said:

but it feels odd to ignore currents in the CPC generated by the system under test, especially in a system that uses the bit of CPC between car and charger  as part of its signalling path.

AMK said:

“currents in the CPC generated by the system under test” Is this matter typically disregarded, or is it simply not given due consideration?

But surely those are between PE and PC between the vehicle and the EVSE only ... not back through the supply PE unless there's an N-PE fault?