A major player in the EVSE market has kindly consented to provide some technical training for the electrical installation tutors at a training centre where I often tutor part-time. I imagine it is not entirely motivated by altruistic considerations but at least it gives an opportunity for the tutors and myself to get some kind of grasp on the various products on offer and where this particular manufacturer sees the direction of travel for EVSE. I am already aware that the company is moving away from products that rely on the installation of earth electrodes in PME situations and are placing considerable focus on load management. Is there any particular question that you would ask them given the opportunity?
Ok Andy, you seem to think that total disconnection if the mains goes 20V down on normal indicates a broken CNE. It may, but may not. You can disconnect everything but disconnecting the Earth connection requires all manner of saftey systems to be safe, a single contactor doesn't do it at all. Remember that Earth connection needs to be disconnected in <40ms, just like an RCD. Any fault will disconnect and reconection cannot be automatic, can it? The fail would be all the time, not even occasionally with a real mains supply. A half-cycle at nearly zero volts is remarkably common and causes no trouble to anything except you mains disconnection idea. The problem is you do not have the time to wait, as waiting could kill someone!
Not quite. BS 7671's requirement is for disconnection for the broken-PEN boxes within 5s not 40ms (722.411.4.1.(iv)) - which might sound poor, but is sort of in alignment with 5s disconnection times for L-PE faults on TN distribution circuits and probably better than for L-PEN faults imported from the DNO's system. I believe the idea is to reduce the likelihood of nuisance tripping due to the L being pulled down by L-N and L-PE faults elsewhere (which can take up to 5s to clear) and should also ride through many other minor events. Yes, I'd agree ideally you'd want something better than a multi-pole contactor for switching the c.p.c. - ideally ensuring it closes first, and have the close verified before closing L & N - and likewise ensuring L & N have opened before opening the c.p.c. - but I'm sure that's doable (safety relays have doing that kind of thing for machine guarding systems for decades) - although BS 7671's requirements on that point don't seem that onerous (543.3.3.101 ((ii) or (iii)).
I guess more of a worry would be an accidental true-earthing of a live conductor (like in that farm thread recently) which can raise the CNE well above true earth without a corresponding drop in L-N voltage. Maybe that's one for the DPC.
Ok Andy, you seem to think that total disconnection if the mains goes 20V down on normal indicates a broken CNE. It may, but may not. You can disconnect everything but disconnecting the Earth connection requires all manner of saftey systems to be safe, a single contactor doesn't do it at all. Remember that Earth connection needs to be disconnected in <40ms, just like an RCD. Any fault will disconnect and reconection cannot be automatic, can it? The fail would be all the time, not even occasionally with a real mains supply. A half-cycle at nearly zero volts is remarkably common and causes no trouble to anything except you mains disconnection idea. The problem is you do not have the time to wait, as waiting could kill someone!
Not quite. BS 7671's requirement is for disconnection for the broken-PEN boxes within 5s not 40ms (722.411.4.1.(iv)) - which might sound poor, but is sort of in alignment with 5s disconnection times for L-PE faults on TN distribution circuits and probably better than for L-PEN faults imported from the DNO's system. I believe the idea is to reduce the likelihood of nuisance tripping due to the L being pulled down by L-N and L-PE faults elsewhere (which can take up to 5s to clear) and should also ride through many other minor events. Yes, I'd agree ideally you'd want something better than a multi-pole contactor for switching the c.p.c. - ideally ensuring it closes first, and have the close verified before closing L & N - and likewise ensuring L & N have opened before opening the c.p.c. - but I'm sure that's doable (safety relays have doing that kind of thing for machine guarding systems for decades) - although BS 7671's requirements on that point don't seem that onerous (543.3.3.101 ((ii) or (iii)).
I guess more of a worry would be an accidental true-earthing of a live conductor (like in that farm thread recently) which can raise the CNE well above true earth without a corresponding drop in L-N voltage. Maybe that's one for the DPC.
