2026 NEC draft - EV and GFCI problems incoming?

[wwhitney]

Nothing in a informational note would be admissible as enforceable in court. The term Class A is probably also unenforceable.

Eh, I think the text we've discussed so far is enough for any reasonable AHJ or jurist to conclude that "Class A" means "UL 489 Class A".

Regardless, in the 2023 NEC 210.8 was changed to require a "listed device". And the definition of "listed" refers to the "appropriate designated standards". Between that and the allowances of Article 90 and also 110.2 which requires equipment to be "approved", any AHJ is on safe ground requiring GFCIs to be UL 489 Class A. As they will do.

Cheers, Wayne

 

[tortuga]

Eh, I think the text we've discussed so far is enough for any reasonable AHJ or jurist to conclude that "Class A" means "UL 489 Class A".

An AHJ would not need a state code amendment for allowing an IEC, CSA, NOM or ETL standard in place of a UL one, if it also provides the equivalent protection to the public, this is done all the time in places that use the NEC, such as Saudi Arabia and the Philippines.
All the breaker manufactures in the US are heavily versed in IEC, and nobody in Europe or the UK is dying because they don't use UL GFCI standards.

And the definition of "listed" refers to the "appropriate designated standards".

We agree that it does not say UL then?
The GFCI situation in the US is a unique mess to this country across several major manufacturing sectors HVAC, home appliances like refrigerators, car manufacturers, and the simple reason is the rest of the world has harmonized on RCD, which in side by side testing also provides the same Class A protection without the nuisance trips.
There is no need to reinvent that wheel we should accept RCD and move on.

 

[wwhitney]

in places that use the NEC, such as Saudi Arabia and the Philippines.

I'm unable to confirm that either Saudi Arabia or the Philippines use an electrical code based on the NEC. If you can provide a link to a foreign country's electrical code that is based on the NEC, and it shows they have not amended the definition of GFCI or 210.8, that would be relevant to the discussion.

Cheers, Wayne

 

[wwhitney]

We agree that it does not say UL then?

I think you are going to have a very hard time convincing any US AHJ that the "appropriate designated standard" is anything other than UL 943, which is referenced by the NEC.

Cheers, Wayne

 

[jim dungar]

An AHJ would not need a state code amendment for allowing an IEC, CSA, NOM or ETL standard in place of a UL one,

As far as I know IEC does not do any testing that results in an independent listing, its standards result in self certification. GFCI is a term used in the UL984 standard which are used by third party labs that are Listing products for the American market.

Definitions matter, GFCI are not the same as RCD or CCID.

 

[tortuga]

I think you are going to have a very hard time convincing any US AHJ that the "appropriate designated standard" is anything other than UL 943,

I bet not as hard of a time as it will be to convince auto manufacturers to care about UL 943.

which is referenced by the NEC.

Not referenced in enforceable language.

 

[Jaybone812]

Correct, page 224 on this link shows it was submitted by William Synder from RCC solutions

https://docinfofiles.nfpa.org/files/AboutTheCodes/70/70_A2025_NEC_P12_FD_PIResponses.pdf

IF he's on this forum I'd love to hear the rational since all that is written on the PI is " The problem is in the interpretation that EVSE when hardwired is required to have class A GFCI protection per 210.8(A)-(F)"

The

Correct, page 224 on this link shows it was submitted by William Synder from RCC solutions

https://docinfofiles.nfpa.org/files/AboutTheCodes/70/70_A2025_NEC_P12_FD_PIResponses.pdf

IF he's on this forum I'd love to hear the rational since all that is written on the PI is " The problem is in the interpretation that EVSE when hardwired is required to have class A GFCI protection per 210.8(A)-(F)"

The PI he submitted for 625.54 led to the rule being revised to require exactly what he was trying to have removed. In his substantiation he said the pi was to provide clarity for people who thought hardwired EVSE required gfci protection.
And it backfired because the proposed revision he wrote all receptacles and outlets . If you read both public inputs for the definition of a individual branch circuit you’ll see why he thought it was to his advantage to include outlets , because both of the public inputs for that definition are ridiculous and one of them says “he never believed a individual branch circuit for a hardwired piece of utilization equipment had a outlet , he said theres not outlet it’s simply a termination obviously they were shut down by the committee and the committees response basically said they didht understand the definition of a outlet , but in a nicer way lol
If you look at those two public inputs as well as one that was put in for motor branch circuits you’ll see clear as day the submittal completely backfired on him

 

[jaggedben]

For what it's worth, Snyder submitted a comment for the Second Draft opposing the inclusion of 'outlets'. Effectively he retracted his original PI for the First Draft. Really unclear to me what he was thinking. I wonder if he'll show his face here again on this issue, he seems to have realized what he did.

There were 21 PIs for this section in the Second Draft. Over half opposed the inclusion of 'outlets', including from Mike Holt and some members of this forum on this thread. Only a couple comments were in support of the inclusion of outlets or trying to make that requirement more explicit; the rest were concerned with other caveats or clarifications (such as adding 20ma 'Special Purpose GFCI' for EVSE supply circuits over 150V to ground). The CMP ignored the majority opinion without substantive comment. FWIW the 3 negative votes all gave good comments: at least the record shows there were good reasons not to adopt this change to the code and it lacks a bona fide argument in favor.

This is going to be bad and I hope states make efforts to edit this requirement out when they adopt the 2026 NEC. Or that AHJs don't enforce it. Otherwise I think customers and contractors will end up removing it anyway after inspection, adding cost and further undermining the legitimacy of the code and NFPA code making process.

You can write your building standards commission or whoever in your state to try to get them to modify this section during their code adoption process. The NFPA is not the final word.

 

[Jaybone812]

For what it's worth, Snyder submitted a comment for the Second Draft opposing the inclusion of 'outlets'. Effectively he retracted his original PI for the First Draft. Really unclear to me what he was thinking. I wonder if he'll show his face here again on this issue, he seems to have realized what he did.

There were 21 PIs for this section in the Second Draft. Over half opposed the inclusion of 'outlets', including from Mike Holt and some members of this forum on this thread. Only a couple comments were in support of the inclusion of outlets or trying to make that requirement more explicit; the rest were concerned with other caveats or clarifications (such as adding 20ma 'Special Purpose GFCI' for EVSE supply circuits over 150V to ground). The CMP ignored the majority opinion without substantive comment. FWIW the 3 negative votes all gave good comments: at least the record shows there were good reasons not to adopt this change to the code and it lacks a bona fide argument in favor.

This is going to be bad and I hope states make efforts to edit this requirement out when they adopt the 2026 NEC. Or that AHJs don't enforce it. Otherwise I think customers and contractors will end up removing it anyway after inspection, adding cost and further undermining the legitimacy of the code and NFPA code making process.

You can write your building standards commission or whoever in your state to try to get them to modify this section during their code adoption process. The NFPA is not the final word.

I did see that comment for the second draft and If was a serious backpedal lol. Because he only submitted the comment as an attempt to save face but it was a failed attempt lol. He definitely realizes what he did but he’ll never admit it and to this day on other social media groups he brags about submitting a PI that passed lol, he said That just last weekend actually on one of his myth of the week posts

 

[jaggedben]

 

[don_resqcapt19]

The language accepted by CMP 12 in the final ballot for the second draft will very likely be subject to a NITMAN and resulting CAM, with extensive debate at the June 2025 NFPA meeting. Changes often result from the floor votes at the annual meeting.

NITMAN Notice of Intent to Make A Motion
CAM Certified Amending Motion.

The language in the final ballot is:

625.54 Ground-Fault Circuit-Interrupter (GFCI) or Special Purpose Ground-Fault Circuit-Interrupter (SPGFCI) Protection for Personnel.
(A) Cord- and Plug-Connected.
(1) 150 Volts or Less to Ground.
All receptacles and outlets rated 150 volts or less to ground, installed for the connection of electric vehicle charging shall GFCI for personnel protection.
(2) Greater than 150 Volts to Ground.
All receptacles rated greater than 150 volts to ground shall have SPGFCI with a ground-fault trip current not exceeding 20 mA for personnel protection.
(B) Permanently Wired.
(1) 150 Volts or Less to Ground.
All outlets rated 150 volts or less to ground, installed for the connection of electric vehicle charging, shall have GFCI for personnel protection.
(2) Greater than 150 Volts to Ground.
All outlets installed for permanently wired electric vehicle charging, shall have SPGFCI with a ground-fault trip current not exceeding 20 mA for personnel protection. This requirement shall become effective January 1, 2029.

Exception No. 1 to (B)(1) and (B)(2): Outlets installed to supply dc charging shall not require SPGFCI protection.

Exception No. 2 to (B)(1) and (B)(2): Outlets installed for electric vehicle bidirectional charging shall not require SPGFCI protection.

 

[Jaybone812]

The language accepted by CMP 12 in the final ballot for the second draft will very likely be subject to a NITMAN and resulting CAM, with extensive debate at the June 2025 NFPA meeting. Changes often result from the floor votes at the annual meeting.

NITMAN Notice of Intent to Make A Motion
CAM Certified Amending Motion.

The language in the final ballot is:

Thank you sir , I heard spgfci was going to be added to 625.54 second draft , thank you for confirming !

 

[tortuga]

SPGFCI, CCID and all is just reinventing a decades long establisehd and proven RCD at a added cost to US manufacturers, consumers and contractors.

 

[Jaybone812]

The language accepted by CMP 12 in the final ballot for the second draft will very likely be subject to a NITMAN and resulting CAM, with extensive debate at the June 2025 NFPA meeting. Changes often result from the floor votes at the annual meeting.

NITMAN Notice of Intent to Make A Motion
CAM Certified Amending Motion.

The language in the final ballot is:

Not a fan of how it’s worded though. My opinion the language used when referring to outlets will lead to confusion . Not that my opinion matters whatsoever lol

 

[don_resqcapt19]

SPGFCI, CCID and all is just reinventing a decades long establisehd and proven RCD at a added cost to US manufacturers, consumers and contractors.

RCD and GFCI are different technical concepts as far as people protection. RCD is intended to prevent ventricular fibrillation and GFCI is based on the let go current. SPGFCI is a bit of a cross between the two.

 

[don_resqcapt19]

Not a fan of how it’s worded though. My opinion the language used when referring to outlets will lead to confusion . Not that my opinion matters whatsoever lol

I made public inputs in an attempt to clear up the confusion related to the term outlet. In all cases, my definition said that the outlet is where the branch circuit conductors are terminated. The inputs for "outlet" and "lighting outlet" were rejected, but the one for "receptacle outlet" was accepted.

Receptacle Outlet.
An outlet where the branch-circuit conductors are connected to one or more receptacles. (CMP 1)

The rejected inputs were:

Outlet.
The point where branch circuit conductors connect to, or will be connected to utilization equipment terminals or conductors. (CMP-1)

Lighting Outlet.
A point where the branch circuit conductors connect to, or are intended to be connected to a lampholder, luminaire, or fixture whip conductors. (CMP-18)

 

[Jaybone812]

I made public inputs in an attempt to clear up the confusion related to the term outlet. In all cases, my definition said that the outlet is where the branch circuit conductors are terminated. The inputs for "outlet" and "lighting outlet" were rejected, but the one for "receptacle outlet" was accepted.


The rejected inputs were:

I actually read all of your public inputs in reference to this and how you used the public inputs made by a certain education to substantiate the pi’s you submitted yourself and enjoyed reading every word of them lol

 

[Jaybone812]

I made public inputs in an attempt to clear up the confusion related to the term outlet. In all cases, my definition said that the outlet is where the branch circuit conductors are terminated. The inputs for "outlet" and "lighting outlet" were rejected, but the one for "receptacle outlet" was accepted.


The rejected inputs were:

And clearly some people do need clarification on rhe definition because or this issue would not even exist !

 

[hillbilly1]

Looks like a better solution if they are hell bent on having a 5 ma trip instead of 20 ma, would just require the manufacturer to do it.

 

[jaggedben]

I made public inputs in an attempt to clear up the confusion related to the term outlet. In all cases, my definition said that the outlet is where the branch circuit conductors are terminated. The inputs for "outlet" and "lighting outlet" were rejected, but the one for "receptacle outlet" was accepted.


The rejected inputs were:
Outlet.
The point where branch circuit conductors connect to, or will be connected to utilization equipment terminals or conductors. (CMP-1)
...

I can see why they rejected this definition since it would make "receptacle outlets" mutually exclusive of "outlets". (There is no place at a receptacle where branch circuit conductors connect to or will be connected to utilization equipment terminals or conductors.) So that would have led to massive additional confusion wherever the code might have a requirement that applied to 'outlets'.

We'll probably end up rehashing previous discussions, but it remains baffling to me how many people seem hellbent on allowing only receptacles to provide GFCI protection at the load end of branch circuit conductors. It makes no practical sense whatsoever. Simply allow any type of listed hardwired equipment to provide GFCI. Then we can stop worrying about where the outlet is.