Receptacle grounded via conduit?

[Fred B]

I think if you know the flex is there you should try to get a compliant EGC there, even if NEC isn't clear about what is required for existing situation.
GFCI may be an acceptable alternative. The problem with flex isn't that it won't carry ground fault current, but it could have enough impedance that overcurrent protection may be less effective and is the main reason for the six foot of flex being max allowed as an EGC path.

NEC (2017) 348.60 - FMC, if flexibility is required after installation a ground conductor is required, but if not the FMC can act as the grounding conductor in accordance with 250.118(5), here it indicates a limited length of FMC otherwise a grounding conductor will be required.
Had an odd thought, providing gfci protection for ungrounded receptacle when replacement with 3 prong, can a gfci breaker provide the same code compliance as a gfci receptacle, or does it need to be a receptacle type?

 

[kwired]

NEC (2017) 348.60 - FMC, if flexibility is required after installation a ground conductor is required, but if not the FMC can act as the grounding conductor in accordance with 250.118(5), here it indicates a limited length of FMC otherwise a grounding conductor will be required.
Had an odd thought, providing gfci protection for ungrounded receptacle when replacement with 3 prong, can a gfci breaker provide the same code compliance as a gfci receptacle, or does it need to be a receptacle type?

The thing here is possibly grandfathering of something that was compliant when installed. I don't know when six foot rule came into play, guessing around same time they figured out AC cable wasn't a very good EGC unless a bonding wire was installed under the sheath, which I think was at least 1950's or 60's.

 

[Fred B]

The thing here is possibly grandfathering of something that was compliant when installed. I don't know when six foot rule came into play, guessing around same time they figured out AC cable wasn't a very good EGC unless a bonding wire was installed under the sheath, which I think was at least 1950's or 60's.

Isn't AC without the bonding strip BX?
AC cable is different than FMC, AC relies on the bonding strip for grounding continuity, FMC is adequate on it's own withing the length limitation. And like BX/MC, AC cannot have added conductor pulled in, FMC like EMT or Ridged can.

 

[kwired]

Isn't AC without the bonding strip BX?
AC cable is different than FMC, AC relies on the bonding strip for grounding continuity, FMC is adequate on it's own withing the length limitation. And like BX/MC, AC cannot have added conductor pulled in, FMC like EMT or Ridged can.

If it is spiral construction same problems cause more resistance as length increases. That bonding strip in the AC cable is to lower overall resistance over whatever the length is by shorting each spiral wrap together so current can follow a shorter straight line so to speak instead of the longer spiral length.

I'd guess AC cable is fine on it's own if only limited to 6 feet as well, but is manufactured in much longer lengths than that and why bother taking the bond strip out just because you maybe can in this case?

 

[Fred B]

If it is spiral construction same problems cause more resistance as length increases. That bonding strip in the AC cable is to lower overall resistance over whatever the length is by shorting each spiral wrap together so current can follow a shorter straight line so to speak instead of the longer spiral length.

I'd guess AC cable is fine on it's own if only limited to 6 feet as well, but is manufactured in much longer lengths than that and why bother taking the bond strip out just because you maybe can in this case?

NEC (2017) 320.108 "Type AC cable shall provide an adequate path for fault current as required by 250.4(A)(5) or (B)(4) to act as an equipment grounding conductor. "
No reference for length limitation.

Wasn't suggesting removing bonding strip, but simply comparing BX and AC. Old BX very much like AC but no bonding strip, and AC, MC, and BX all are fabricated assembly, not like FMC.

 

[kwired]

NEC (2017) 320.108 "Type AC cable shall provide an adequate path for fault current as required by 250.4(A)(5) or (B)(4) to act as an equipment grounding conductor. "
No reference for length limitation.

Wasn't suggesting removing bonding strip, but simply comparing BX and AC. Old BX very much like AC but no bonding strip, and AC, MC, and BX all are fabricated assembly, not like FMC.

I'm suggesting the fact the conductors are already inside is the reason. The sheath itself is pretty much about the same thing as FMC and very possibly is pretty effective EGC at six foot lengths, but just isn't made anymore without the bond strip already in it so why even need to worry about such 6 foot wording in the code likely part of why it isn't there?

There is no limit on length of FMC but there is limit on how much length can be utilized as EGC, and my best guess is for about the same reasons that AC cable requires the bonding strip.

I have read stories of people with old AC cable that seen heated up cable sheath or signs of it being heated before because of fault current on the sheath. It can add enough resistance to the fault return path to either retard overcurrent device operation or possibly even not allow operation at all.

 

[don_resqcapt19]

...
Had an odd thought, providing gfci protection for ungrounded receptacle when replacement with 3 prong, can a gfci breaker provide the same code compliance as a gfci receptacle, or does it need to be a receptacle type?

406.4(D)(2)(c) permits upstream GFCI protection. However, no matter where the protection is installed, don't forget to read Informational Note No, 2. That sends you to a list of equipment that is not permitted to be plugged into any receptacle that does not have an actual equipment grounding conductor.

 

[Fred B]

406.4(D)(2)(c) permits upstream GFCI protection. However, no matter where the protection is installed, don't forget to read Informational Note No, 2. That sends you to a list of equipment that is not permitted to be plugged into any receptacle that does not have an actual equipment grounding conductor.

Yes

 

[kwired]

406.4(D)(2)(c) permits upstream GFCI protection. However, no matter where the protection is installed, don't forget to read Informational Note No, 2. That sends you to a list of equipment that is not permitted to be plugged into any receptacle that does not have an actual equipment grounding conductor.

There is code, but then there is actual physics. Even if not code compliant I'd much rather see the GFCI used here than just letting it be. Best protection would be GFCI on the supply end instead of the load end though.

Same goes for old 12 AWG NM cable that had the 16 AWG EGC in it.

 

[don_resqcapt19]

There is code, but then there is actual physics. Even if not code compliant I'd much rather see the GFCI used here than just letting it be. Best protection would be GFCI on the supply end instead of the load end though.

Same goes for old 12 AWG NM cable that had the 16 AWG EGC in it.

I agree and tried to add an exception for the 2023 code to say that 250.114 does not apply where replacement receptacles are installed per 406.4(D), but CMP 5 rejected it and from the discussion I listened from their First Draft meeting, the panel statement will say that a GFCI is no a substitute for an EGC for the items listed in 250.114. I believe adding the GFCI without an EGC would be an improvement over leaving the old two wire receptacle. Maybe some Public Comments when the First Draft is published for comment would change the minds of the panel members.

 

[kwired]

I agree and tried to add an exception for the 2023 code to say that 250.114 does not apply where replacement receptacles are installed per 406.4(D), but CMP 5 rejected it and from the discussion I listened from their First Draft meeting, the panel statement will say that a GFCI is no a substitute for an EGC for the items listed in 250.114. I believe adding the GFCI without an EGC would be an improvement over leaving the old two wire receptacle. Maybe some Public Comments when the First Draft is published for comment would change the minds of the panel members.

Maybe this sort of thing depends on which group lobbies the CMP the hardest - the GFCI manufacturers or the copper industry.

 

[don_resqcapt19]

Maybe this sort of thing depends on which group lobbies the CMP the hardest - the GFCI manufacturers or the copper industry.

Most of the discussion by CMP 5 was related to the listing standards. Equipment that has a grounding prong on the cord is only listed for use on a circuit that has an EGC. Adding GFCI protection does not change the listing standard requirement.

 

[kwired]

Most of the discussion by CMP 5 was related to the listing standards. Equipment that has a grounding prong on the cord is only listed for use on a circuit that has an EGC. Adding GFCI protection does not change the listing standard requirement.

But yet we have been allowed to replace two wire non grounding receptacles for many years now with GFCI protected grounding receptacles, as long as the intended load isn't something mentioned in 250.114, which covers many common items that would have a grounding conductor in the supply cord.

 

[tortuga]

Regarding old cables, I have done some research on this for renovation projects.
The NEC used to have tables for sizing raceway and conduit being used as an EGC.
From my notes:
They had grounding type outlets way back to the 1920's they just were not code required.
There were rules in the NEC around what an EGC was and how it was sized.
Old versions of the NEC going way back (rule 907-a of the 1930 NEC) had some wording like:
"The size of the grounding conductor for conduit, cable sheath or armor, and other metal raceways or enclosures for conductors, and for equipment, shall not be less than given in Table 250-95." (1965 wording)

There was never a size under 1/2 inch allowed to be used as an EGC.

The 1965 NEC had two tables one for 'Sizes of conductors for grounding interior raceways and equipment' and 250.95(b) 'Sizes of Equipment Grounding Circuit Conductors for a grounded system' , both tables required a raceway or cable armor to be sized for use as an ecg, no sizes under 1/2IN trade size were ever allowed.
So if you had some large BX cable in 1/2 Armor say #10's or 8's? the sheath could be used as the EGC.

As of the 1959 NEC 334-2(a) required Type AC to have a bonding strip. “ Cables of the AC type, except ACL,shall have an internal bonding strip of copper or aluminum, in intimate contact with the armor for its entire length.” Type ACL was a lead covered cable and the lead served as the bonding strip, but AC cable would not be allowed to be used for equipment grounding until the 1968 NEC when
250-95 was reworded to only apply to aluminum and copper equipment grounding conductors.
Using raceway or cable armor was moved to 250-57 allowing all sizes of AC cable in 1968.

So in my opinion from historic research, If back in 1951 the original installer say wanted to install a 3 prong grounding type receptacle (for whatever reason) on that box and it was fed from the 1/2" flex; code would have allowed the flex to act a the EGC,
If the 3/8 bx was the 'home run' it would not have been allowed.

 

[tortuga]

Here is a phot of the table from the 1951 NEC

 

[don_resqcapt19]

But yet we have been allowed to replace two wire non grounding receptacles for many years now with GFCI protected grounding receptacles, as long as the intended load isn't something mentioned in 250.114, which covers many common items that would have a grounding conductor in the supply cord.

And we have never been permitted to use the items listed in 250.114 where there is no physical EGC at the receptacvle

Here is a phot of the table from the 1951 NECView attachment 2554773

I think those wire sizes were in place until the late 60s. I was working in a local hardware store while in high school, and, some time while I was working there, the size of the EGC in NM changed.

 

[kwired]

And we have never been permitted to use the items listed in 250.114 where there is no physical EGC at the receptacvle

Guess my point is unless it is a receptacle intended for a specific appliance, how can it be enforced?

So if you have one in a spot in a kitchen that is obviously intended for a refrigerator, an inspector can reject it. But that don't mean user can't place a fridge someplace else and plug into what otherwise wasn't intended for a refrigerator, and have no EGC there.