Cor, yes, you are 100% correct, but there are also many shitty EVSEs out there that aren't "well designed" or even UL listed. They are all over Amazon, and people buy more of them every day.
As proof, I have several that were given to me that trip the upstream GFCI as soon as they are connected, without even needing to connect an EV! (Assumably the POST is causing it, but I did not verify) Regardless, I still advise hardwire for all daily use EVSEs, as this is all too common: https://ingineerix.com/pic/?melty1450 A capacitive dropper might be safer than a resistor for leakage current detection, but could also just use the microcontroller to generate a LF AC signal into a separate test coil (wound more turns). That would be able to precisely generate a repeatable test current which doesn't depend on the incoming AC. For a recommended wall-mount EVSE, I now steer people to Tesla's well-designed combo unit that does 48A and both NACS and J1772: https://shop.tesla.com/product/universal-wall-connector On Fri, Apr 12, 2024 at 11:17 PM Cor van de Water via EV <ev@lists.evdl.org> wrote: > Phil, > NOPE! The *correct* designed EVSE will have a test circuit that indeed > generates a GFCI test fault current but *only* in the EVSE's GFI > circuit, not in the upstream 240V connection. > The reason that an EVSE often trips the upstream GFCI is not due to > the EVSE but due to the *huge* capacitance in the EV charger inlet > which can easily cause enough fault current, just like many lab > equipment has large filter capacitors on their AC input to filter out > EMC, but at the same time tripping the GFCI to the point that you > cannot even leave some bench power supplies plugged into the wall, > even turned off, because their AC filters will happily cause more than > 5mA current through their heavy capacitors and nuisance-trip the > breaker. > > For the EVSE to test their built-in GFCI protection of the > *downstream* charging cord, all that is needed is that the test > circuit connects a wire (and resistor) across the 2 phases of the 240V > so that there is *no* upstream fault current, and then route the test > wire through the GFCI detection Current Transformer, so that a one-way > fault current is introduced of between 20-30mA, while verifying that > the detection circuit indeed trips on the application of the test > current. (The GFCI Current Transformer has both phase wires of the > charging cord routed through it, so that any difference of more than > 20mA between the two wires - even at 40A or more of charging current - > will cause a trip of the GFCI detector and disconnect the relay > contacts that feed the 240V input to the charging cord.) > In the JuiceBox charging station I only used a 4mA test current, so I > could use a smaller and cheaper limit resistor, by looping the test > wire 5x through the GFCI CT coil, so the 4mA was counted 5 times to > yield the 20mA test current injection to verify the correct operation > of the GFCI after *every* charging session, as well as during > boot-selftest. You might notice that the JuiceBox (at least the > plastic JuiceBox 2) is also UL approved for operation on 110V and it > was a challenge to make the GFCI test circuit still generate a 20mA > test current no matter what voltage that the EVSE received. If you > figure it out, let me know what you found and I will tell you if you > got it right. > Cor. > > On Fri, Apr 12, 2024 at 12:14 PM (-Phil-) via EV <ev@lists.evdl.org> > wrote: > > > > IIRC, J1772 specs 20ma. So it will likely put an over 20ma ground fault > > intentionally during its self test. > > > > On Fri, Apr 12, 2024 at 12:01 PM EV List Lackey via EV < > ev@lists.evdl.org> > > wrote: > > > > > On 12 Apr 2024 at 13:35, Mark Hanson via EV wrote: > > > > > > > An HTML attachment was scrubbed.. > > > > > > Thanks Phil etc > > > > > > Sometimes the code doesn't explain why there's certain rules that > appear > > > arbitrary (like the 12" cord rule). The larger units are limited to > 6'. > > > > > > Presumably the GFCI trip current (inside a EVSE) is higher (not sure > how > > > much) than the 6ma (if using a GFCI panel breaker). A large chassis > like > > > a > > > car and with an internal switching supply-charger has EMI X (line to > line) > > > and Y caps (line to ground for common mode RF) that will most likely > leak > > > close to 6ma. Luckily inspector 13 (here) hasn't enforced the panel > GFCI > > > breaker requirement (since 2020) on 14/50s used for EV charging. > > > > > > Best regards Mark > > > > > > PS from the moderator - please try to send plain text, folks. Thanks. > > > > > > 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 > > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > > > > > The phrase "May you live in interesting times" is the lowest in > > > a trilogy of Chinese curses that continues "May you come to the > > > attention of those in authority," and finishes with "May the > > > gods give you everything you ask for." I have no idea about > > > its authenticity. > > > > > > -- Terry Pratchett > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > > > > > _______________________________________________ > > > 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: < > http://lists.evdl.org/private.cgi/ev-evdl.org/attachments/20240412/68fe2494/attachment.htm > > > > _______________________________________________ > > 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/ > > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.evdl.org/private.cgi/ev-evdl.org/attachments/20240413/5cdbf616/attachment.htm> _______________________________________________ Address messages to ev@lists.evdl.org No other addresses in TO and CC fields HELP: http://www.evdl.org/help/
