-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Ground fault protection a new complication for EV charging?
- Thread starter Plugsuvohio
- Start date
It’s required in the 2017 Code.
https://www.nfpa.org/assets/files/AboutTheCodes/70/Proposed_TIA_1242_NFPA_70.pdf
Yes, but that is inside the electronics box and then covers the rest of the way down the cord to the car. It is not protecting upstream, where you are grasping the plug and pushing it into the outlet. So that is where the new NEC is requiring that outlets installed for EV charging have to have GFCI breakers.
All in it seems like a permanently installed EV charger is a reasonable price and has substantially better safety than continued use of the mobile charger / EV charger with a NEMA style connector.
I personally plan on swapping out my (outdoor) nema 6-20 plug in a weather resistant box for a clippercreek LCS-20; when I got my car I asked for (and was given!) a second J1772 adapter that I was thinking I'd permanently lock into a non-tesla branded EVSE.
Installing one of these would require "just" replacing the receptacle with a plate, running the pigtail into the box in a weather-safe manner, wire-nutting this thing's pigtail into place, and properly attaching this thing to the wall?
In such an installation, there would no longer need to be GFCI for the circuit that runs from the EVSE to the main panel, correct?
I personally plan on swapping out my (outdoor) nema 6-20 plug in a weather resistant box for a clippercreek LCS-20; when I got my car I asked for (and was given!) a second J1772 adapter that I was thinking I'd permanently lock into a non-tesla branded EVSE.
Installing one of these would require "just" replacing the receptacle with a plate, running the pigtail into the box in a weather-safe manner, wire-nutting this thing's pigtail into place, and properly attaching this thing to the wall?
In such an installation, there would no longer need to be GFCI for the circuit that runs from the EVSE to the main panel, correct?
davewill
Active Member
I would agree that for permanent outdoor use, best to use something truly weatherproof, like the LCS-20. Your installation description sound like it covers all the bases. One thing I'd suggest is having the pigtail loop down then back up into the electric box, so that water doesn't run straight down the cable into/onto the box. Probably not necessary, but...
Hm; if I wanted to be truly belt+ suspenders, pull the romex back into the house, poke the pigtail through the wall, and put all that mess into another box in the basement, and possibly have a loop / sag in the pigtail to prevent water intrusion into the inside box that actually has the electrical junction?
I assume that it is kosher to feed an EVSE using a 2 wire circuit (I'm reusing a "legacy" nema 5-20 circuit where I had 12 gauge romex with "hot/neutral+ground" and I had it rewired to be "hot/hot+ground" (and colored tape on the hot-neutral wire)?
davewill
Active Member
Yeah, Hot-Hot-Ground is fine for a hardwired EVSE. Since the LCS-20 comes with a nice armored pigtail already attached and properly terminated inside the unit, I'd just go with running it into your weatherproof box, and using wire nuts to connect right there, like it shows in the manual:
https://store.clippercreek.com/index.php?route=product/product/download&download_id=12
That's just from being on these forums too much.
I just talked to an inspector. No. He said,"...unless the charger has a built in GFCI protection..." So, no. As long as it is GFCI protected, either in the unit or from the breaker or from a GFCI outlet, it's protected.
All EVSE's (including the UMC and Wall Connector) have built in GFCI protection; it's part of the J1772 spec, so that's sort of a moot point.
If you're on NEC 2017, any 120V or 240V outlet, 50A or less, installed for the purpose of EV charging requires a GFCI, per 625.54:
625.54 Ground-Fault Circuit-Interrupter Protection for Personnel. All single-phase receptacles installed for the connection of electric vehicle charging that are rated 150 volts to ground or less, and 50 amperes or less shall have ground-fault circuit-interrupter protection for personnel.
Note the receptacle must have GFCI protection. Downstream GFCI protection in the EVSE plugged into the receptacle doesn't cut it. The inspector may not flag a code violation under 2017 if the required GFCI is missing from the outlet, but they don't have the authority to waive portions of the code if it's been adopted.
Now I suppose you could argue that the receptacle was installed for a purpose other than EV charging. I don't know if that would exempt you. My city is on 2011, so I'm as familiar with the later versions of the NEC.
JPoldo
Member
Let's assume some owners never remove the UMC plug from receptacle as Tesla recommends. Consider what occurs when there is a ground fault at receptacle. The UMC is in series so it too would have no ground and recognize a fault. Therefore, I believe the user sees a ground fault. At that point, I would turn off circuit breaker and locate/repair the problem.
For what its worth, I tried my Gen 1 UMC on a 120 volt GFI outlet at a cottage back in August. The UMC turned on but would not deliver power to the car. The UMC light flashed a fault code. I tried cycling the GFI outlet with the test/reset sequence but still no power. The GFI looked like it was operating normally based on the LED indicator. I routed the UMC cable into the cottage and tried a non-GFI outlet. The UMC worked fine. I didn't have any tools to check the GFI even for reversed hot and neutral lines but I don't think it would have shown a normal LED indication if they were reversed. My conclusion is that the Gen1 UMC can have problems if plugged into a GFI outlet.
davewill
Active Member
That sounds more like a wiring problem than a GFI problem. The thing to do (next time it happens) is to count the number of flashes so you can find out what the error actually is.
kkillebrew
Banned
I can't see where a properly installed 14R-50 with GFI would do anything other than increase safety. Should have no effect on draw, charging, or voltage.
It has been explained elsewhere, and it would be just as long for me to go search for it as to just re-type the explanation here of what the problem is.
The way the Tesla UMC tests the ground connection is to try to run a very tiny bit of current between a hot and ground to make sure the ground side is solidly holding to 0V instead of floating. That particular action of taking current from one hot and not sending it back through the other hot is exactly the kind of thing that GFCI is supposed to detect and prevent. However, it is supposed to have a threshold where a very tiny amount won't trip it, so the Tesla UMC should be small enough to get by. But a very large portion of GFCI outlets and breakers tend to degrade with age, so that their threshold levels drift and don't work right, and the ground test from the Tesla UMC will set them off pretty frequently.
So while you say you think it would only increase safety, it also has a very common side effect of causing a lot of false nuisance breaker trips.
kkillebrew
Banned
I said it SHOULD increase safety. Sorry you are having the problems, I understand your frustrations and the situation now. Seems this should be on Tesla to fix... but that does not solve your problem though.
There's nothing for Tesla to "fix." The leak test @Rocky_H described is part of the J1772 spec. If anything needs to be "fixed," it's the arguably boneheaded requirement for redundant GFCI's in 2017 625.54.
JPoldo
Member
While installing a NEMA 14-50 receptacle in my garage, I tried the 120V, 20 amp GFCI protected outlet with the UMC. The GFCI breaker is located in a panel at least 75 ft away. Numerous loads such a high power leaf blower, power washer, battery charger, etc have worked fine in this outlet. However, within 20 seconds of charge using the UMC (8/2017 vintage), my breaker trips. There are no other loads on this 6 outlet circuit. I'm a bit suspicious of cascaded GFCI's. Probably the GFCI needs replacing, but there's no incentive now since the NEMA 14-50 is operational, delivering 29 mi/hr.
Similar threads
