I need to run two new 20A circuits to the other side of the house - at Home Depot I see this "12-2-2" romex cable with 2 hots, 2 neutrals, and one ground. Is it safe to share that one 12ga. ground between two 20A circuits? (I "assume" it was a 12 ga ground, it's up in the air on the spools at HD so hard to get a close look)
Does 12/4 armored cable (MC or BX) only have one grounding conductor? or is it better because the metal jacket is an additional/supplemental grounding path?
Thanks.
Interesting; I had never heard of 12-2-2. But reading up on it, it is specifically intended for 2 AFCI protected circuits. Since AFCIs also function as GFCIs, any current going to the ground will trip the breaker and cut off the current. There is no possible way to overload the ground.
Without the AFCI you could, in principle, overload the ground without tripping the breaker so it would not meet code. Personally, I would rather run two cables anyhow.
At least this is my reading of it; I can't wait to hear how foolish I am...
Not [necessarily] true. *Some* (perhaps most) AFCIs are dual-listed for use as GFCIs as well. But not all of them.
And the ones that are, don't function quite the same: the trip threshold is a bit higher.
Not true.
Very unlikely, I grant, but not impossible: a low-impedance (hence high current) short from hot to ground on *each* hot leg would indeed overload the ground. This would not necessarily involve an arc that would trip an AFCI. It
*is* extremely unlikely, but it's not impossible.Even *with* the AFCI, you could, in principle, overload the ground without tripping the breaker, as I noted above.
Unless it's a really, really long run, it's probably a *lot* less expensive to use two 12-2 WG cables, and two standard breakers. AFCIs ain't cheap.
And it's a lot harder to screw up.
-- Regards, Doug Miller (alphageek at milmac dot com)
Nobody ever left footprints in the sands of time by sitting on his butt. And who wants to leave buttprints in the sands of time?
IMHO:
Since the ECG only carries current for a very short time, in a ground fault, you should have no problems sharing the ground. I've run emt with many circuits and a single ground wire. So, imho, I don't know of any problems with the code against doing this.
BTW, now if you are interestinged in running two circuts in the same cable, why have you ruled out 12/3 ?
Personally, I like having a seperate ground, so this rules out AC as my choice. Plus, as for the MC jacket, it can only be used as a ground if listed as such. Eventhought it is grounded.
From your questions, you are doing the right thing planning out the project before you runn around buying stuff.
Might want to post what you doing(or intend on doing) to get suggestions how to do it best.
hth,
tom @
AFCI's do not perform as GFCI's unless they are specifilcy built as such and labeled. This is not the general case.
How would you overload the ground, if you follow the NEC?
Learning everyday,
tom
You couldn't, since you wouldn't be using 12-2-2. But he is talking about using 12-2-2.
Just to be clear, there is no requirement to have an AFCI protect a
12-2-2 cable. The point is that one application for 12-2-2 is running two circuits to back to back bedrooms; since the circuits must be AFCI protected, they can not share a neutral.Cheers, Wayne
Certainly. But without a AFCI the two circuits would have to be on opposite legs because of the shared ground, and then there is no advantage of 12-2-2 over 12-3. In fact 12-3 is better because it reduces the voltage drop. So while 12-2-2 can be used without AFCI, it would not make sense to do so.
Why do you believe that to be the case, can you indicate a section of the NEC that requires it? I thought that if you are running two circuits in conduit, for example, it was standard to only pull one EGC. Is there a requirement to have a separate EGC for each circuit?
Admittedly it is a speciality product, but I can see other places it would make sense. For example, in a kitchen setting with two small appliance circuits, protected by separate GFCI outlets, you need to separate the neutrals downstream of the GFCIs.
Cheers, Wayne
Utter nonsense. If they shared a *neutral* they'd need to be on opposite legs.
Please don't give any more electrical advice until you figure out the difference between ground and neutral.
It does what?
-- Regards, Doug Miller (alphageek at milmac dot com)
Nobody ever left footprints in the sands of time by sitting on his butt. And who wants to leave buttprints in the sands of time?
You might be right; I couldn't find anything on it. However, it seems prudent to have a ground that can carry the maximum current supplied by the hots.
Yes, it could be used for that, but I would rather run two cables.
It does seem true that sharing a neutral between the two circuits on separate legs reduces the voltage drop due to the wiring. The current on the neutral leg is less than on either hot, so the voltage drop due to the neutral leg is less than it would be with a separate neutral.
Cheers, Wayne
Careful, people might confuse 30 ma GFCI trip as GFCI protection for personnel. So, imho, best not to mention afci/gfci unless the item is labeled and listed as such.
later,
tom (VERY PARINOID) ;)
Irrelevant. This would do nothing to alter the voltage drop between the supply and the load, which is the only place it really matters. Further, the resistance of whatever load is applied to the circuit, even if it's just a single light bulb, is orders of magnitude greater than the resistance of the conductors.
-- Regards, Doug Miller (alphageek at milmac dot com)
Nobody ever left footprints in the sands of time by sitting on his butt. And who wants to leave buttprints in the sands of time?
I don't believe that is correct. Given a fixed voltage available at the panel, and for example a resistive load, the hot and neutral wires are effectively resistors in series with the load. The total resistance of the circuit will be the resistance of the hot + resistance of the load + resistance of the neutral. You can't ignore the neutral just because it is after the load.
Another way of looking at things is that using a shared neutral makes the overall circuit partially parallel, and two resistors in parallel have a lower resistance than they do in series.
Well, this is always true, yet on a long run conductors are typically oversized to avoid an excessive voltage drop. So it does happen. Admittedly the benefit of reduced voltage drop due to a shared neutral may be more theoretical than of practical importance, but I do believe it is real.
Cheers, Wayne
I had it in my head that sharing a neutral for two circuits was a big no-no, so I never considered that option....
As it turns out, I found a better route through the house and have enlisted a helper, so I'm going to KISS and run two 12/2 romex cables...
What we're doing:
Circuit #1 - 20A - To support an electric hydronic baseboard heater, 750W. Also on the circuit will be under tile electric warming mat at about 300W. A GFCI breaker in the panel to protect any faults with the baseboard heater.
Circuit # 2 - 20A - For general lighting and outlets in the bath. GFCI outlets of course.
I decided to run two circuits because I couldn't see one circuit handling baseboard heater + tile warming mat + lights + fan + hot curling iron +
1500W hair dryer.... Sure I could probably get away with it until she kicks the hair dryer into overdrive ...-- Paul
Very good, I agree with this. Voltage drop due to the conductors only matters for long runs. But it is still a reasonable question whether sharing a neutral between two circuits will reduce that voltage drop, even if it only matters for long runs.
True, the voltage drop between the source and the load only depends on the resistance of the hot. But the "available voltage" at the load depends on the voltage drop of both the hot and the neutral.
Consider an example: exactly 120V available at the panel, a single circuit with (long) wiring with a resistance of 0.1 ohms each for the hot and the neutral, and an electric heater with a resistance of 10 ohms (nominally a 1200W heater). If the heater were wired directly to the panel, the total resistance is 10 ohms and the heater would draw
12 amps. With the (long) wiring, the total resistance is 10.2 ohms, and the heater draws 11.76 amps (120/10.2). The voltage drop across the hot is 1.176 volts, the drop across the heater is 117.6 volts, and the drop across the neutral is 1.176 volts.Now suppose we have two 10 ohm heaters on an Edison circuit (shared neutral). This represents the best case of totally balanced loads, and the current on the neutral will be 0. So this is equivalent to having the two heaters in series across a 240V supply. Each hot conductor has the same resistance of 0.1 ohms. The total resistance is 20.2 ohms, and the current is 11.88 amps (240/20.2). The voltage drop across the first hot is 1.188 volts, the drop across each heater is 118.8 volts, and the drop across the second hot is again 1.188 volts.
In the single circuit case, the "voltage drop" the single heater saw was 2.4 volts (120 - 117.6). In the shared neutral case, the voltage drop was only 1.2 volts (120 - 118.8). This illustrates the best case scenario of using a shared neutral: with perfectly balanced loads, the voltage drop due to the wiring with be half as much.
Cheers, Wayne
I believe (b) above is not always required by the NEC, although it seems like a good idea for a variety of reasons. 210.4(B) requires this when the circuit supplies "more than one device or equipment on the same yoke". And 210.4(C) requires this if the circuit serves any
240V loads.Cheers, Wayne
Yes, *if* you have perfectly balanced, purely resistive loads.
But the difference has no practical significance in any case.
-- Regards, Doug Miller (alphageek at milmac dot com)
Nobody ever left footprints in the sands of time by sitting on his butt. And who wants to leave buttprints in the sands of time?
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.