Bx Cable (AC)

Dennis Alwon

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
If you come across 12/2 BX cable that doesn't have that steel wire then we know it cannot be used as a ground. However, it is inherently already grounded and can carry a fault. The code will not allow us to use the cable as a grounding conductor. If I install a gfci and then install receptacles with a ground then am I not, inherently, using the ground?

The ground on the receptacle is connected thru the screws that attach the receptacle to the box and then to the cable. Indirectly are we creating a hazard by allowing 3 wire receptacles on a BX cable without the inner steel wire?
 
Dennis Alwon said:
If you come across 12/2 BX cable that doesn't have that steel wire then we know it cannot be used as a ground.
Click to expand...
There's some debate about this because when the cable was manufactured wasn't the outer jacket considered an EGC just like FMC in lengths greater than 6'. That muddies the water when you want to install a 3-wire receptacle because the NEC is silent on the issue. The GFCI protection with the 3-wire receptacle should be fine and is not prohibited by the NEC.

Personally I think that the problem of the AC overheating during a fault is possible but it is somewhat overblown. Yes it can happen but around here there are thousands of old homes that all have this type of cable still in use. I haven't seen any evidence to show that they're rapidly going up in flames.
 
LarryFine said:
But, interconnecting the EGC terminals of such GFCI-protected receptacles is.
Click to expand...
There is no wire type EGC in Dennis' example so there is no connection other than the mounting screws connecting the receptacle to the metal box which is connected to the armor.
 
The gfi will know the differance between the yoke connected (according to you) to bonded jacket and the actuall ground term...
Will not work- definately will not work w/a gfi recept tester unless you do silly things...
 
rambojoe said:
Will not work- definately will not work w/a gfi recept tester unless you do silly things...
Click to expand...
Why wouldn't it work? The GFCI receptacle IS actually grounded through the cable armor. The question is since that cable armor is no longer an acceptable EGC is it code compliant to install the 3-wire receptacle?
 
rambojoe said:
The gfi will know the differance between the yoke connected (according to you) to bonded jacket and the actuall ground term...
Will not work- definately will not work w/a gfi recept tester unless you do silly things...
Click to expand...
A GFCI doesn't know, or care, if there is an EGC connected. It works off of any imbalance between hot and neutral current.
 
Little Bill said:
A GFCI doesn't know, or care, if there is an EGC connected. It works off of any imbalance between hot and neutral current.
Click to expand...
I think that Joe was commenting about using a receptacle tester which simulates the imbalance by sending it to the EGC.
 
I think it's important to point out to the owners of property with the old steel BX can be a fire hazard as the jacket can heat to a dangerous level when it carries a ground fault.
 
rambojoe said:
If memory serves (gasp) it wont reset.
Click to expand...
It won't trip without a ground however with the screws of the gfci into the metal box may indeed allow it to trip. This may be the saving grace.

If an appliance is plugged in and there is a short to ground then it is possible the overcurrent protective device won't open but the gfci probably will.
 
I think the issue is overblown.

From what I have read bx with the added ground strip was required around 1959 when the switch was made to 3 wire receptacles.

My thinking is when you look in the code at the armored cable requirements did they ever have 2 articles? One for Bx with the ground strip and one without the ground strip/ No not that I am aware of. Both types are armored cable.

I have heard around here that without the strip it was referred to as "rat proof Romex"

I agree with @infinity that the code appears to be silent on this and I to suspect that at the time the older cable was made it was probably approved as an equipment ground. Would like to know if that was true. The information must be available somewhere. Maybe from UL

My house which I sold 4 years ago was all wired with BX originally and was built in 1955. It had both types of BX cable in it some with the ground strip and some without.

The older type bx without the strip (at least in my case) had heavier steel armor which was pretty substantial.

If connected properly I don't have much doubt about it carrying a fault
 
Eddie702 said:
The older type bx without the strip (at least in my case) had heavier steel armor which was pretty substantial
Click to expand...
My house was built in 1955 too and had the same situation. I did see that all the #12 BX did have the ground strip. But then there was no such thing as AC lite with aluminum armor back then. I know we all saw the picture of the old BX armor glowing red and I wonder what the OCP was.

-Hal
 
Would be interested in some quantified information about this. E.g. what is the 60 Hz AC resistance per foot of old BX armor? #14 Cu is 3.1 ohms per kft.

Cheers, Wayne
 
@wwhitney The post below is from an old Mike Holt forum post. If he truly meant 3 ohms/100' then that not so good

old bx cable

Hello The armored cable without a ground wire maybe from the 1940s is that 16 gage wire? I here it's grandfathered but most likely on new 15 amp breakers with half house refurbishes. Could someone please tell me the ratings of the old armored cables from the 40s? Thanks
forums.mikeholt.com

Until 1959, it may have been a 16 or 18 awg al, or 20 cu, if there at all. As of 1959 the bonding strip was required. 1959:334-2(b)

Some tests done in 2007 to old AC showed that the insulation samples all passed a 5000 volt test, but the resistance of the sheath tended to be high, maybe over 3 ohms per 100 feet, I think.
Click to expand...
 
You must log in or register to reply here.
Top