NATIONAL FIRE PROTECTION ASSOCIATION - NFPA 70, Tentative Interim Amendment (TIA)

[mtnelect]

Ground-fault circuit-interrupter protection shall not be required for listed
HVAC equipment. This exception shall expire September 1, 2026.

 

[tortuga]

I love this part:

justification of the action.
Almost every state that has adopted the 2020 Edition of the NEC have modified or deleted
Section 210.8(F).

I dont know why no one in the US likes the idea of using ma trip levels like they do in the rest of the world.

 

[mtnelect]

I love this part:

I dont know why no one in the US likes the idea of using ma trip levels like they do in the rest of the world.

The NFPA has formed a Task Force to find out the problem of "Phantom Tripping" with DC conversion equipment, which is causing problems with HVAC units. Stay tuned for their findings. In the meantime the NFPA has given them until 2026 to find out.

 

[ActionDave]

I love this part:

I dont know why no one in the US likes the idea of using ma trip levels like they do in the rest of the world.

I agree. The attitude of, "If it's not GFCI then someone is going to die" makes no sense to me. We could have less nuisance tripping, greater compliance, and safer electrical installs if GFPE level protection was expanded instead of requiring GFCIs everywhere they added in the last code cycle.

 

[infinity]

It's more likely that a sick elderly person will die if their AC unit nuisance trips during a heat wave than someone dying from touching the unit. This is a horrible code change which indicates that some of the CMP were asleep at the wheel.

 

[tortuga]

I love this quote in there also:

Evidence of safe use in Japan with a different means of protection has
been documented.

They can add the UK, France, Germany just about every other country that uses RCD's or what we call GFPE.
I am pretty sure the guts of these HVAC units are the same world wide.
All the major breaker manufacturers here make the class "b" gfci breakers so its not like they need to make a new breaker.

 

[ActionDave]

It's more likely that a sick elderly person will die if their AC unit nuisance trips during a heat wave than someone dying from touching the unit. This is a horrible code change which indicates that some of the CMP were asleep at the wheel.

I love this quote in there also:

They can add the UK, France, Germany just about every other country that uses RCD's or what we call GFPE.
I am pretty sure the guts of these HVAC units are the same world wide.
All the major breaker manufacturers here make the class "b" gfci breakers so its not like they need to make a new breaker.

It's the all or nothing mentality that is pervasive in the culture of the code writing members that is the problem. We can't use GFPE level of protection because it's not safe enough compared to GFCI level protection even though it would be easier to implement and a huge improvement in electrical safety. It's aggravating and shameful.

 

[don_resqcapt19]

It's the all or nothing mentality that is pervasive in the culture of the code writing members that is the problem. We can't use GFPE level of protection because it's not safe enough compared to GFCI level protection even though it would be easier to implement and a huge improvement in electrical safety. It's aggravating and shameful.

The real difference between GFPE and SPGFCI is the required time to trip. GPFE does not have any requirements for how long between the detection and a ground fault that the device opens. Special Purpose GFCIs (SPGFCI) do have a specified time to trip.

 

[ActionDave]

The real difference between GFPE and SPGFCI is the required time to trip. GPFE does not have any requirements for how long between the detection and a ground fault that the device opens. Special Purpose GFCIs (SPGFCI) do have a specified time to trip.

I thought the difference was the 5mA vs a higher level of ground fault current the device would sense before it opens.

 

[don_resqcapt19]

I thought the difference was the 5mA vs a higher level of ground fault current the device would sense before it opens.

That is part of it, but the standard has no time to trip requirements for GFPE. SPGFCIs operate in the same 20 to 30 mA trip as many GFPEs, but even the trip current level is not specified for GFPE in the product standards.
As seen with the RCDs used in other countries, a 30 mA trip can protect people with a quick enough trip time. Those devices are typically required to trip with in 250 milliseconds. A Class A GFCI is permitted to take up to 1.78 seconds to trip with a 30 mA ground fault, per UL 943.

 

[ActionDave]

Either way, the incident that led to the code change in 2020 where a missing equipment ground on an air conditioning unit caused a death could have been prevented if a GFPE breaker that we already are allowed use for heat tape had been installed, and if it were required now instead of GFCI protection we would be better off. As it stands the NEC loses credibility and we are right back to where we were before the 2020 code.

 

[tortuga]

As seen with the RCDs used in other countries, a 30 mA trip can protect people with a quick enough trip time.

Thats the point I think the CMP is missing.
We should just adopt one of the RCD standards in the NEC for this then UL wont need to charge us all for re writing something thats been solved for almost 30 years now.

 

[kwired]

RCD's used in many other countries are on systems with ~240 volts to ground, so to be fair that maybe makes some difference in the needed trip threshold necessary vs a system with 120 volts to ground.

That said class A GFCI should still be limited primarily to cord and plug connected items and even more so to items utilizing 5-15 and 5-20 type receptacle/plugs. You almost never see a missing EGC pin on a cord cap for anything else. I don't even have an issue with 30 mA protection for hard wired art 680 applications. Receptacles for portable pools and such maybe class A GFCI is still good idea.