Just installed the Tesla wall version 3 and manual states integrated GFCI and no additional required [CCID20]
Also in bold letters: do not install a GFCI circuit breaker. Unit was hardwired in the garage.
hard wired EV chargers
- Thread starter Treg the electrician
- Start date
Also in bold letters: do not install a GFCI circuit breaker. Unit was hardwired in the garage.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
Yes, CCID20, not Class A GFCI. So the use of the word GFCI in that statement is not accordance with the NEC definitions.
That instruction translates to "you may not install this product in a manner that would cause the NEC to require a GFCI. Choose a different product if you want to do such an installation."
Cheers, Wayne
Does the NEC require a GFCI on any 60A circuit? I can’t think of any.
jaggedben
Senior Member
- Location
- Northern California
- Occupation
- Solar and Energy Storage Installer
Not yet.
electrofelon
Senior Member
- Location
- Cherry Valley NY, Seattle, WA
Ya know that would be a perfect new NEC requirement. No history or data of a problem, and they can implement it when no products are available and screw everyone over. I'm going to make some calls to the CMP's to make sure they thought of this one.
- Location
- Illinois
- Occupation
- retired electrician
Since we were discussing residential, I hadn’t considered 3 phase. Thanks.
- Location
- Illinois
- Occupation
- retired electrician
680.22(A)(4) would require GFCI protection for a 60 amp receptacle within 20' of a dwelling unit pool in the 2023 code.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
Definitely a good example for the question in post #23, but I think worth noting that in the context of the OP, 625.44 requires EVSEs over 50A to be hardwired.
Cheers,
Wayne
- Location
- Illinois
- Occupation
- retired electrician
Both 625.44(A) and (B) permit the use of 125/250 volt, single phase, 60 amp receptacles. It appears that is a 2023 change.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
I should have known I'd get in trouble just checking the 2020 NEC. : - )
I'm curious why the 2023 NEC 625.44 only allows 60A receptacles for 125/250V 3 pole grounding, not 250V 2 pole grounding. Does the equivalent of NEMA 6-60 not exist, only the equivalent of NEMA 14-60?
Thanks,
Wayne
The problem is no one wants to make stuff for just the US market anymore. I know there is a lot of empirical data to back up our 5ma GFCI, but one has to wonder why the rest of the wold has settled on 20 and even 30 ma as their 'gfci'?
I wonder if there is a movement to get at least the IEC / EU to adopt the 5ma GFCI definition then manufacturers would not be so resistant to support it
I wonder if there is a movement to get at least the IEC / EU to adopt the 5ma GFCI definition then manufacturers would not be so resistant to support it
- Location
- Illinois
- Occupation
- retired electrician
No idea why. The Public Input which was rejected and the Public Comment which resulted in the code change were submitted by one ChargePoint.
- Location
- Illinois
- Occupation
- retired electrician
The difference is the higher limit is set at a point where the current does not cause ventricular fibrillation and our lower current is based on 95% of the adult population being able to let go of the conducive object.
There is also a difference in the permitted time to trip between the two standards. UL 943 would permit the GFCI to have a trip time of 4.5 seconds for a 7 mA ground fault, while the IEC standards require a trip time of 1/4 second at their rated trip current. The permitted trip time for a 30 mA fault on a UL 943 GFCI is 0.58 seconds.
Note that most GFCIs trip much faster than the maximum permitted trip time...if they didn't, you would have to hold the test button in for over 5 seconds.
Interesting thanks, I appreciate how knowledgeable you are on this topic. I am typically in favor of the IEC standards over UL (such as 508) and a was a proponent of harmonizing the two (glad thats mostly done),
But in the case of wet locations I think the EVSE manufacturers should just bite the bullet and integrate a 5ma GFCI into their 'outlets'.
jaggedben
Senior Member
- Location
- Northern California
- Occupation
- Solar and Energy Storage Installer
I keep wondering if there's an engineering difficulty in limiting leakage current to 5ma at higher EVSE amperages. Otherwise why would the manufacturers object to using a GFCI breaker. I second that standards should be harmonized internationally and wonder why CCID20 shouldn't be good enough for the US.
I suspect it may be related to way they detect the absence of an EGC.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
FWIW, the EVSE UL standard covers measuring and limiting ground leakage current, which I assume would also cover during the ground assurance testing of the EVSE. Although I didn't check the limits, I would assume they are below the 4-6 ma GFCI trip threshold.
But if the EVSE puts any current on the EGC, even only 1 or 2 ma, as part of its ground assurance test, then in conjunction with capacitive leakage current on the branch circuit, that could trip a GFCI breaker.
On the other hand, if the EVSE implements CCID5, and if 210.8(F) is interpreted so that the outlet is the vehicle connector itself (what you plug into the car), and the EVSE is hardwired, then there would be no need for upstream GFCI. The question then is if how often the EVSE itself would see more than 5 ma of leakage current from the J1772 cord and the EV, during normal operation.
Cheers, Wayne