Neutral In Switchbox For Lighting Circuit

[stanleyk]

A two gang switchbox has a 3 way switch for one lght and an on-off switch for a 2nd light.
One three conductor cable comes into the box for the 3 way lighting circuit.
A second three conductor cable enters the box for the on-off switch. This cable carries the extra neutral conductor required by NEC 404.2(C.)
Both lighting circuits are powered by the same source.
I don't question that 404.2(C) is satisfied, but how does this situation mesh with article 300.3(B)?
I maintain that 300.3(B) is not violated so long as the cable covering, as well as the boxes at both the switch and fixture locations, are non magnetic.

 

[Dennis Alwon]

IMO, since the cable is non metallic it does satisfy code however it will create unwanted emf's in the event that the neutral is used for the 3 way in the future.

 

[dicklaxt]

I agree not a problem...

Dennis,I don't understand the unwanted EMF's , not being picky but I don't see a problem.

dick

 

[RustyShackleford]

A related question, not threadjacking I hope ...

Suppose a multi-gang box has switches controlling multiple lighting loads that are powered by more than one circuit. Is it sufficient to provide a single neutral conductor to the box, or must there be a neutral from each of the multiple circuits (whose switch legs are present in the box) ?

From my reading it does not seem to be a code requirement. But if "fancy" switches requiring un-switched power (the reason for the new code requirement, as I understand it) were added to more than one circuit, but only one neutral exists in the box, then there could be the situation of sharing a neutral between two different circuits (if the multiple fancy switches received hot from multiple circuits but shared the one neutral).

 

[fmtjfw]

A related question, not threadjacking I hope ...

Suppose a multi-gang box has switches controlling multiple lighting loads that are powered by more than one circuit. Is it sufficient to provide a single neutral conductor to the box, or must there be a neutral from each of the multiple circuits (whose switch legs are present in the box) ?

From my reading it does not seem to be a code requirement. But if "fancy" switches requiring un-switched power (the reason for the new code requirement, as I understand it) were added to more than one circuit, but only one neutral exists in the box, then there could be the situation of sharing a neutral between two different circuits (if the multiple fancy switches received hot from multiple circuits but shared the one neutral).

2014 will most likely require a neutral for each circuit. You could probably get away with a single neutral if all the hots are part of a MWBC. That would include adjacent Breakers and handle ties.

The point of the neutrals is to get away from the practice of using the grounding conductor as a neutral. This was done when occupancy sensors first came out. The IBEW got a change to require the neutral because all the "few" milliamps per sensor added up in commercial settings to shocks for workers who were working with grounding conductors.

The change to explicitly require the "right" neutral is from a more general principle which will require neutrals to be identified with their associated hot conductor(s) when more than one is present in a box ...

It turns out that identifying a neutral with its hots is a difficult problem without markings.

 

[RustyShackleford]

The point of the neutrals is to get away from the practice of using the grounding conductor as a neutral. This was done when occupancy sensors first came out.

I've never installed an occupancy sensor. Thought I'd check out a typical one:

http://www.lutron.com/TechnicalDocumentLibrary/0301605.pdf

... sure looks to me like the manufacturer is saying to use grounding conductor as a neutral.

It turns out that identifying a neutral with its hots is a difficult problem without markings.

No doubt, but requiring them to be in the same jacket of NM would help a lot.

 

[RustyShackleford]

Here's another one:

http://www.leviton.com/OA_HTML/ProductDetail.jsp?partnumber=ODS0D-IDT&section=38557&minisite=10251

... and it specifically states "Neutral Wire Connection: Not Required".

Similarly, it turns out that many of the Lutron sensors, including the one I just cited, do not require a neutral either:

http://www.lutron.com/TechnicalDocumentLibrary/048469.pdf

Looks like the NEC has given us a solution to a problem that does not exist. Or exists only if the person installing the switch is incompetent and/or lazy and the switch is of the type that requires a neutral (does this perhaps mean a discontinued or obsolete model ?).

 

[jumper]

IIRC UL will be changing its standards so that the use of the grounding conductor as a neutral for these sensors will no longer be allowed.

 

[don_resqcapt19]

I've never installed an occupancy sensor. Thought I'd check out a typical one:

http://www.lutron.com/TechnicalDocumentLibrary/0301605.pdf

... sure looks to me like the manufacturer is saying to use grounding conductor as a neutral.


....

That is the exact reason for the code rule. The use of the EGC as the neutral for the occupancy sensor is a shock hazard if someone would open the EGC between the device and the power source. When you open the EGC, you will have 120 volts between the two conductors. The current UL standard limits the current to 0.5 mA, and that amount of current is not a serious shock hazard, it is enough to get your attention and your reaction to the shock may result in you falling off a ladder or something like that. Also the current is additive and each device can draw up to the 1/2 mA.
The code rule was needed because UL said they would not change the product standard to prohibit the use of the EGC as the grounded conductor for the devices unless the code was changed to require a grounded conductor at the switch locations.

 

[fmtjfw]

I've never installed an occupancy sensor. Thought I'd check out a typical one:

http://www.lutron.com/TechnicalDocumentLibrary/0301605.pdf

... sure looks to me like the manufacturer is saying to use grounding conductor as a neutral.


No doubt, but requiring them to be in the same jacket of NM would help a lot.

The problem in residential wired in cable is relatively simple. The problem is in commercial and industrial settings with multiple neutrals in a single conduit. for example if you have 3 neutrals in a single conduit and 6 hots which appear in multiple junction boxes and you pick the neutral to use with a hot "randomly" there is a very good chance that you could overload the neutral.

I never came up with a good way to identify, using meters or circuit tracers, the neutral -- hot groups without turning off the power and lifting neutrals from the bar.

 

[RustyShackleford]

The thread that won't die, but it's one of the most irksome 2011 changes. Revisiting the OP's question about 300.3(B) problems with running the 404.2(C) neutral for a lighting switch, but with a slightly different circuit ... I hope I can explain it well enough without a drawing.

I'm doing two-location switching, but one of the 3-way switches is dead-end. Why am I doing this, even though it's perfectly feasible to run line to one switch and load to the other, and run neutral between them (on the 3-conductor wire with the travelers) ? Because the load is the light fixture of a ceiling fan, so I need hot there where the 3-conductor cable to the fan/light starts; so the cable to the fan/light will start where the non-deadend switch and line are.

But as it turns out, I will have neutral, but not hot (of the same branch circuit) at the box with the dead-end switch, via another piece of cable. It's actually another pair of 3-way switches wired in the usual neutral-with-travelers way.

There's no issue of unbalanced currents for the main loads - the wiring is perfectly normal pre-2011. But if an occupancy sensor switch were installed on the dead-end switch, the return current would flow in the neutral in that other cable. Of course the unbalanced current could only be 1/2ma, so unlikely EMI or induction heating would be a real issue.
Also, the two 3-conductor cables (for the two pairs of 3-way switches) are more or less "together" (they are loosely bundled traveling along a stud wall through drilled holes). I'd hate to have to run a 4-conductor cable for this.

 

[RustyShackleford]

But if an occupancy sensor switch were installed on the dead-end switch, the return current would flow in the neutral in that other cable. Of course the unbalanced current could only be 1/2ma

Ok, it WAS a pretty silly concern I guess. I'm an electrical engineer in my other life ...