For 2014: Neutrals and Switches

[renosteinke]

So the 2011 will require the presence of a neutral at switches. I see two 'trickle down' effects to this.

The first is that someone is sure to start adding a 'dummy' screw for landing the neutral - and, before you know it, someone will be saying that you're required to use the dummy screw, or tie all the neutrals together, etc.

The second applies to those devices - illuminated switches, photocells, motion sensors, timers, etc. - that use the ground wire as their 'neutral.' (Keep in mind that only SOME do this - not all!). Someone is going to next want these devices to be 'banned,' and require a neutral terminal on them.

The second brings up the interesting wording of another code section, which bans "objectionable" current over the ground wire. This is the 'loophole' tht has allowed these devices to operate until now. I can't see it changing.

I can't see that changing, because I cannot find the words to write such a rule. You can't just say 'fault current only,' simply because you still need the ground to carry transients from some surge suppressors, frequency drives, and all manner of electronics. Besides, how would you prevent RF-induced currents without putting everything in pipe?

I start this thread because NOW is the time to start thinking this through - rather than be surprised in 2014 with another "done deal."

 

[480sparky]

.............The first is that someone is sure to start adding a 'dummy' screw for landing the neutral - and, before you know it, someone will be saying that you're required to use the dummy screw, or tie all the neutrals together, etc...........

I don't see that happening at all. Why would the manufacturers retool their plants for something that's not required?

 

[winnie]

IMHO adding this as a section of the NEC was an error. This is an issue that should have been corrected with a change to the appropriate UL standards.

I pretty much stated my opinion in this thread:
http://forums.mikeholt.com/showthread.php?t=119537

IMHO it is quite reasonable to permit say 50uA of current to 'leak' from a switch _intentionally_ in order to power the switch. This is in the same range as the current that you would expect to 'leak' to ground simply because of capacitive coupling in 100 feet of romex. 50uA at 120V is 6mW, enough to do some reasonable control work with good design.

With the 500uA allowed under recent standards, a dozen devices would be enough to trip a GFCI and present a noticeable (but not large) shock hazard. At 50uA per device, you would overload the circuit well before you had problematic leakage.

-Jon

 

[iwire]

With the 500uA allowed under recent standards, a dozen devices would be enough to trip a GFCI and present a noticeable (but not large) shock hazard. At 50uA per device, you would overload the circuit well before you had problematic leakage.

As an electrician I do not want to start treating EGCs as live nor do I want to get surprised by one. Only to have someone at a comfy desk saying 'it was not that bad'.

I am against the code change but I am also against the EGC being used as a circuit conductor.

 

[Dennis Alwon]

I suspect the problem is with commercial jobs where there are many motion sensors, timers etc that utilize the ground. I personally don't see the change as bad but I agree it would be easier to make the manufacturer design a device that didn't need the ground--- we all know the mfg. are in on the code panels so.. without bringing politics any further we will leave it at that.

 

[480sparky]

I suspect the problem is with commercial jobs where there are many motion sensors, timers etc that utilize the ground. I personally don't see the change as bad but I agree it would be easier to make the manufacturer design a device that didn't need the ground--- we all know the mfg. are in on the code panels so.. without bringing politics any further we will leave it at that.

Resi has a lot of stuff that goes in switch boxes that need a noodle. It's been a proposal to require it in the switchbox for many many code cycles. I guess the CMP finally got tired of reading them and accepted it just to shut 'em up.

 

[nakulak]

Allowing current on the egc threatens everything the egc is supposed to provide in the way of safety. If you allow devices to supply current on the egc, who is to say how many of the devices will be installed ? So if you have 50 apartments with 2 of the devices in each one, you now have potentially 100 times that small current, potentially flowing on an egc somewhere near the gear, maybe it's the central laundry room where everything is made of metal, and now that current is flowing (normally) through the washing machines or any other bonded metal appliance providing a path for someone to be harmed. BAD IDEA.

 

[don_resqcapt19]

Allowing current on the egc threatens everything the egc is supposed to provide in the way of safety. If you allow devices to supply current on the egc, who is to say how many of the devices will be installed ? So if you have 50 apartments with 2 of the devices in each one, you now have potentially 100 times that small current, potentially flowing on an egc somewhere near the gear, maybe it's the central laundry room where everything is made of metal, and now that current is flowing (normally) through the washing machines or any other bonded metal appliance providing a path for someone to be harmed. BAD IDEA.

It will not be flowing on the items that are bonded to the EGC. It will be flowing from the device on the EGC to the main bonding jumper and then to the grounded conductor. Yes, current will will energize everything connected to the EGC, but the voltage that will energize these items is the voltage drop caused by the current flow on the EGC. The current levels are so low that it would be difficult to measure this voltage drop. This voltage drop is also the only thing that would drive current through you if you are touching an item connected to the EGC and some ground object that is not bonded to the EGC. Again, a value so low that it would be difficult to measure. It will not be additive on the EGCs until they connect to the grounding bus at the main panel.

The real safety issue is you between an open EGC that is carrying this current and some grounded object. While the current itself is too low to cause harm, the reaction to the shock may very well cause harm.

That being said I don't think it was a good idea to permit this use and that was done by UL. I think they are behind the new code rule and expect that their standards will be updated to prohibit the use of the EGC as a current carrying conductor.

 

[Dennis Alwon]

Resi has a lot of stuff that goes in switch boxes that need a noodle.

I agree . Many electronic dimmers and other types of switches need a neutral but they don't use the egc unless you do an illegal install. This is why I almost always have neutrals in my switches.

 

[jumper]

I am also against the EGC being used as a circuit conductor.

Allowing current on the egc threatens everything the egc is supposed to provide in the way .

What they said.

 

[dbuckley]

The first is that someone is sure to start adding a 'dummy' screw for landing the neutral...

Here in New Zealand it is custom and practice (though not a regulatory requirement) to run neutrals to switches, and as it happens our switches have a "spare" terminal to gather the neutrals...

 

[winnie]

In general I agree that the EGC should not be used as a current carrying conductor.

However the EGC _always_ carries a bit of current, if only unintentionally. There is _always_ some leakage and _always_ some capacitive coupling.

I really think that it is reasonable to permit intentional current flow to the EGC, if it is of such low levels as to be equivalent to the unintentional current that _must be expected_ in an ordinary installation.

A 20 foot switch loop will have aout 100 foot run romex will have about 800pF of capacitance between the 'hot' and the EGC (this is a very rough number; I see references to 15pF per foot for hot to neutral, and figure slightly higher hot to ground). This means that for a 120V circuit you would expect about 35uA of current just from capacitive coupling alone, with absolutely no galvanic leakage across the insulation.

Thus in ordinary installations, you are probably already dealing with several hundred uA of current flowing on the EGC. Adding a little bit to this _intentionally_ won't change the safety of the install, as long as we define 'a little bit' to be in line with what is already there and _must_ already be there as part of the physics of the situation.

-Jon

 

[winnie]

A 20 foot switch loop will have aout 100 foot run romex will have about 800pF

Please read the above as 'A 20 foot switch loop will have about 800pF '

-Jon

 

[barclayd]

I think UL made a grievous error in listing equipment that intentionally uses the Equipment Grounding Conductor to carry current.
I think the Code-Making Panel erred in adding 404.2(C).
Two proposals are needed for 2014 -
1. Re-define "Equipment Grounding Conductor" in 100.
2. Add another exception that would exempt switches used for porch lights, patio lights, and any other location where a possible, future, maybe some day switch-box-mounted occupancy sensor installation would be totally absurd.
I don't do residential, so I don't really have a dog in this fight other than being opposed to stupidity (other's, not my own) and regulational meddling.
db

 

[brother]

I think UL made a grievous error in listing equipment that intentionally uses the Equipment Grounding Conductor to carry current.
I think the Code-Making Panel erred in adding 404.2(C).
Two proposals are needed for 2014 -
1. Re-define "Equipment Grounding Conductor" in 100.
2. Add another exception that would exempt switches used for porch lights, patio lights, and any other location where a possible, future, maybe some day switch-box-mounted occupancy sensor installation would be totally absurd.
I don't do residential, so I don't really have a dog in this fight other than being opposed to stupidity (other's, not my own) and regulational meddling.
db

What I have highlighted is the MAIN issue here. The egc should NEVER intentionally be used as a ccc during normal operations no matter how small the current is. It takes away on what it was orginally designed to do.

 

[tom baker]

The manufacturers got around the issue of using the ground as a neutral by removing the OFF from the motion sensor. When OFF all power was disconnected as in a switch. So with no OFF there was still the leakage current and its enough to startle you if you get in series with it.

 

[macmikeman]

Here is my guess. By the year 2014 there probably will be an abundance of wireless switch devices that need no connection to the system wiring at all. They already have Lutron Auro Ra for instance which is wireless dimmer switches. Power will still need to be run to light fixtures. But switch wiring is going to be dumbed down for average joe homeowner so he can just stick on a wireless switch anyplace and just set dipswitches and he's in business. Forget the dipswitches, he will just use his phone to set it into the system..... Wired residential light switches are going to go into the dinosaur bin.

 

[Dennis Alwon]

Here is my guess. By the year 2014 there probably will be an abundance of wireless switch devices that need no connection to the system wiring at all. They already have Lutron Auro Ra for instance which is wireless dimmer switches. Power will still need to be run to light fixtures. But switch wiring is going to be dumbed down for average joe homeowner so he can just stick on a wireless switch anyplace and just set dipswitches and he's in business. Forget the dipswitches, he will just use his phone to set it into the system..... Wired residential light switches are going to go into the dinosaur bin.

Mac, I don't understand your post. Auro Ra needs system wiring to work and I believe the newer units need the neutral. I just install an older model Maestro wireless where no neutral was needed. The new models now require a neutral. I believe there must have been a lot of pressure on the industry to make switches with a neutral rather than use the egc. Now the NEC has to catch up. Here is another example of the Nec being a design manual when it insists it is not. I think it is a good change but....

 

[DTLight]

There were also quite a few switches in residential word that used the lamp as their connection to neutral when the fixture was off, and the voltage drop through the triac when it was on to power the switch guts. Switches of this design frequently no longer work when they are powering CFL's or other lamps with power supplies because these higher efficiency lamps do not have a low impedance when they are off.

 

[jwelectric]

I think it is a simple case of people not wanting to call a switch an outlet so the NEC mandated a neutral be installed to settle the argument. Now there is no question that a switch is an outlet.

Now tell me you don't think that was funny