Lee, I have designed EVSE so you can quote me on this: The GFI limit that (level2, AC) EV Charging Stations are designed to for the USA is 20mA. (For EU the limit is 30mA, the same limit as the standard GFCI breakers in EU, although there is an additional requirement to detect 6mA DC Current leakage) Apparently the car manufacturers recognised that 5mA is not a good limit as even 20mA can cause nuisance trips, I have dealt with a few installations of our chargers where the GFI alerted on a problem, despite nothing being wrong with the charging station. The common denominator that I have seen is noise on the grid. In one installation it was perfectly reproducible - during the day, not at night. That set me on the path of Solar being the cause and indeed, that was a home that had recently installed a *lot* of solar panels, each with its own micro-inverter. And I found out that the solar was at the same phase as the EV charger (this was in EU, with 3 phase power, so the customer could move his single phase charger to a different phase than the solar and his problems ceased). Apparently the high frequent switching of the inverters leaked into the phase with high enough power that the heavy EMC filtering caps in the EV caused the charger to see a "leakage" to ground on the phase which was absent on the Neutral wire, hence the nuisance trips of the GFI circuit in the charger. NOTE that GFI in an EV charging is a *very* safe circuit, because it is required for safety certification that the GFI circuit is verified to detect a fault current after *every* charging session or on a regular schedule, I believe once a day. So, every charger with safety marking has a "fault-generating" circuit built-in that causes a 20mA fault current through the Current Transformer that detect the GFI. Usually this is a separate wire from the high power wire, specifically to induce the fault current without affecting the charging cable.
On Mon, Mar 11, 2024 at 12:43 PM Lee Hart via EV <ev@lists.evdl.org> wrote: > > Cor van de Water wrote: > > my garage has a GFCI circuit. As a consequence, I cannot use > > my garage to do any development work, because as soon as > > I plug in one of my HP power supplies, even before I turn it on, > > the heavy capacitive filtering on the AC line will trip the GFCI > > because it indeed creates a current to ground. > > The easy answer is a big isolation transformer. I have some big 1KW ones that > I use. (If anyone needs one, I have many, holding down the basement floor). > > > If I am not mistaken, the NEC has a class of GFCI (and I used to have > > a breaker) that trips at 50mA, I believe there is even a 500mA limit. > > The original idea for a GFCI was safety; to keep people from being > electrocuted (i.e. junior sticks a fork in the toaster with one hand to get > his bread out, while the other hand is resting on the grounded sink. The 5mA > limit was chosen as the maximum current that a normal healthy person would > survive, since they can let go and jump away from the source of the shock. In > medical settings, an even lower limit of 0.5mA was required, on the basis > that hospital patients may well not be able to remove themselves from the > source of the shock. > > But these limits proved to be hard to design for. So manufacturers have > lobbied for higher limits of 50mA. That's more than enough to kill a person. > Yet the last I knew, our EVs are being designed to allow up to 50mA of ground > leakage before their GFCI trips. > > 500mA for a GFCI limit? Gah! That would not only kill someone, but *cook* > them as well! > -- > Excellence does not require perfection. -- Henry James > -- > Lee A. Hart https://www.sunrise-ev.com > > If I am not mistaken, the NEC has a class of GFCI (and I used to have > a breaker) that trips at 50mA, I believe there is even a 500mA limit. > This is only used in industrial settings where a lower limit will > indeed trip guaranteed. > > On Mon, Mar 11, 2024 at 10:50 AM (-Phil-) via EV wrote: > > > > Keep in mind that excepting North America, only (part of) Japan uses a > > lower voltage. In the US (residential) system, no conductor is ever over > > about 160v peak-to-peak with respect to ground, whereas in NZ/EU you are > > getting over 300v P-P, which is arguably 4 times more lethal. I'd > > definitely want everything protected by GFCI/RCD if I had those voltages > > everywhere. > > > > On Mon, Mar 11, 2024 at 10:25 AM EV List Lackey via EV > > wrote: > > > > > On 10 Mar 2024 at 23:41, (-Phil-) via EV wrote: > > > > > > > Based on what I know, [the US NEC is] one of the most rigorous codes in > > > > the world. > > > > > > Agreed. I've seen some ... interesting ... wiring practices elsewhere, > > > including Spain, Italy, France, Canary Islands, Puerto Rico, and South > > > Korea. > > > > > > Some of them look like old USA practices. Example: junction boxes aren't > > > usually used for surface mounted luminaires in France. The cable or smurf > > > tubing emerges from the ceiling or wall. > > > > > > I've seen single conductors run through ceramic cleats on the ceiling > > > surface in South Korea, similar to early 20th century US wiring. It > > > appeared > > > to be a recent installation. > > > > > > Service capacities are also lower. A typical western EU service will be > > > 6kW > > > or 12kW, a size the US hasn't seen in probably 70 years. Spain has a lot > > > of > > > 3kW services. I'm sure that that's a problem for EV home charging there. > > > > > > On the other hand, as Bill says about NZ, in most (all?) western EU > > > nations, > > > the whole house is GFI (RCD) protected at 30ma leakage current. > > > > > > David Roden, EVDL moderator & general lackey > > > > > > To reach me, don't reply to this message; I won't get it. Use my > > > offlist address here : http://evdl.org/help/index.html#supt > > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > > > > > Interpreter: One who enables two persons of different languages > > > to understand each other by repeating to each what it would have > > > been to the interpreter's advantage for the other to have said. > > > > > > -- Ambrose Bierce, "The Devil's Dictionary" > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > > > > > _______________________________________________ > > > Address messages to ev@lists.evdl.org > > > No other addresses in TO and CC fields > > > HELP: http://www.evdl.org/help/ > > > > > > > > -------------- next part -------------- > > An HTML attachment was scrubbed... > > URL: > > _______________________________________________ > > Address messages to ev@lists.evdl.org > > No other addresses in TO and CC fields > > HELP: http://www.evdl.org/help/ > > > _______________________________________________ > Address messages to ev@lists.evdl.org > No other addresses in TO and CC fields > HELP: http://www.evdl.org/help/ > > > > _______________________________________________ > Address messages to ev@lists.evdl.org > No other addresses in TO and CC fields > HELP: http://www.evdl.org/help/ > _______________________________________________ Address messages to ev@lists.evdl.org No other addresses in TO and CC fields HELP: http://www.evdl.org/help/
