Can I use flexible metallic conduit in a run of EMT ?

[Dennis Alwon]

Correct me if I’m wrong but I think what @Dennis Alwon is driving at is the ability of the flex to carry fault current back to a wire type EGC (or pipe) to actually carry fault current back to the source. For example if there’s a 500’ run of 2” FMC, the NEC doesn’t permit it to be used as an EGC. However, if there’s a fault halfway through the run to the raceway then the FMC still needs to act like an EGC in a sense to carry fault current to the appropriate EGC. Comparing it to a piece of PVC in the middle of the run is a different animal, PVC isn’t conductive and won’t carry any current anywhere.

That is exactly my point.

Well, for one thing, if the flex is bonded to an EGC at both ends then faupt current can travel on both directions on the flex.

You are still depending on the flex to carry the fault. I realize this is not an issue in real life situation- or, at least, I don't think it is, however why does the nec say that flex is not an equipment grounding conductor and yet they expect it to carry ground fault current and it probably will.

 

[Fred B]

Code allows for the use of flex (FMC) as grounding conductor within limitations per 348.60.
2023NEC Offers additional guidance in that same section.

I would think that if you are using FMC in a manner you described it would require a separate EGC. One way to not have to run entire length of EGC is a bonding jumper, then you would just need to bond BOTH ends at the EMT connection point, then your EMT otherwise would continue to serve as your EGC. 250.102(D),(E)(2) and Sized per 250.122

 

[ggunn]

..what happens between 1 1/4” and 1 1/2” that stops FMC from being a suitable EGC, even though it still might need to carry fault current?

Could it have something to do with the increase in the length of the metal making up the spiral in the conduit as the diameter increases?

 

[NoahsArc]

Probably because the larger sizes (usually) house larger conductors of greater currents.

Surface area of copper increases exponentially with size, while diameter of pipe only linearly increases. This could make sense.
It always seemed odd to me that given flex is pretty continuous throughout, both as a continuous strip of metal and that metal is slotted into itself throughout. I don't see how it's that much worse than EMT vs a small wire EGC.

Well, for one thing, if the flex is bonded to an EGC at both ends then faupt current can travel on both directions on the flex.

So if it's a 30ft piece of flex, but you can only use flex for 6ft as an EGC, what happens if there's a fault directly in the middle? Sure you get double the paths, but if you're more than quadruple the 6ft rating you still have a problem... I suppose this isn't the biggest concern and most faults are going to occur at outlets and in pull boxes.

If one used boxes instead of transition couplers then one would only have to install the wire EGC from box to box.

This probably is necessary anyway if it's a funky run of flex.

Even without boxes, surely you could just run a bonding jumper between the two sections of EMT around the FMC?

What's the equipment for this? I don't think I've ever seen it done like that in the field.

 

[Elect117]

348.60 - "Where flexibility is not required after installation, FMC shall be permitted to be used as an equipment grounding conductor when installed in accordance with 250.118(5)."

"According to ANSI/UL 1, Standard for Flexible Metal Conduit, FMC longer than 6 feet has not been judged to be suitable for grounding purposes. If the length of the total ground-fault return path exceeds 6 feet or the circuit overcurrent protection exceeds 20 amperes, a separate EGC must be installed with the circuit conductors according to 250.118(5)." - NEC 2020 Handbook

Where vibrations are the purpose of using flex then it can not be used as a EGC bonding equipment because the vibrations can loosen it's connection to the equipment over time.

Where vibrations or the loosening of the equipment connections is not a concern, it can be a EGC for a short length of 6'

The equipment grounding conductor is used to bond equipment. When FMC doesn't qualify as a EGC then it is just conductive equipment that is required to be bonded to the equipment grounding conductor (250.134).

A fault within the conduit would either melt the insulation of the wires or use it's bonded connection to the EGC to clear. This is basically the same for EMT, but they test EMT and it's connectors for sustaining fault current.

I think it is less that the FMC wouldn't carry fault current, and more that they don't know if it can sustain the current long enough for the OCPD to operate.

 

[LarryFine]

"According to ANSI/UL 1, Standard for Flexible Metal Conduit, FMC longer than 6 feet has not been judged to be suitable for grounding purposes. If the length of the total ground-fault return path exceeds 6 feet or the circuit overcurrent protection exceeds 20 amperes, a separate EGC must be installed with the circuit conductors according to 250.118(5)." - NEC 2020 Handbook

Thinking literally, the bold text does not say that the length of the flex alone, but that the entire ground-fault path, cannot exceed six feet, which includes the rigid sections of conduit, too.

 

[Elect117]

Thinking literally, the bold text does not say that the length of the flex alone, but that the entire ground-fault path, cannot exceed six feet. Doesn't that include the rigid sections of conduit, too?

"The combined length of flexible metal conduit, flexible metallic tubing, and liquidtight flexible metal conduit in the same effective ground-fault current path does not exceed 1.8 m (6 ft)."

The language pulled direct from 250.118(5)(d) sounds more like 6' of FMC total for a fault path. Not so much that the total conduit being used as a fault path must be 6' or less when combined with a different EGC conduit method. The language from the handbook is misleading when taken literally.

 

[Choice_Gorilla]

Could it have something to do with the increase in the length of the metal making up the spiral in the conduit as the diameter increases?

Maybe. Assuming the fault current path is the spiral, and not the whole diameter like a stick of pipe, maybe there isn’t a great enough increase in the size of the ribbon that makes the spiral in the larger sizes. Basically the cross sectional area of a piece of the ribbon doesn’t scale appropriately as the size of the FMC increases. Totally speculating of course.

 

[Ohm2]

My point is that the NEC is saying that the flex is not sufficient as a ground so they say to add an equipment grounding conductor. How does adding an equipment grounding conductor (Insulated) make the flex capable of carrying the fault if the NEC says it is not a ground.

Because the EGC is a dedicated, reliable path designed to carry fault current, even if it’s insulated, the fault current will travel through the EGC to the electrical panel’s ground bus. The EGC doesn’t need to be uninsulated to perform this function because electricity will still flow along the path of least resistance, and the EGC provides this path. Once the fault current is significant enough, it will cause the circuit breaker to trip.

 

[electrofelon]

Because the EGC is a dedicated, reliable path designed to carry fault current, even if it’s insulated, the fault current will travel through the EGC to the electrical panel’s ground bus. The EGC doesn’t need to be uninsulated to perform this function because electricity will still flow along the path of least resistance, and the EGC provides this path. Once the fault current is significant enough, it will cause the circuit breaker to trip.

Electricity follows ALL paths back to the source.

 

[Ohm2]

Electricity follows ALL paths back to the source.

Electricity follows ALL paths back to the source.

Correct! No debate there. But was being specific to the question EGC. But you are 100% correct

 

[electrofelon]

Correct! No debate there. But was being specific to the question EGC. But you are 100% correct

What Dennis was saying is that a wire EGC will do nothing for a fault IN and TO the flex. The fault will still need to travel entirely along the flex on its way back to the source.

IMO, logically, all raceways should be acceptable as an EGC. How can it not be an acceptable EGC but yet have to be an EGC for a fault to itself??