On 10/24/2013 10:19 AM, Scott Lurndal wrote:
> Nate Nagel <
njn...@roosters.net> writes:
>> On 10/23/2013 03:02 PM, Scott Lurndal wrote:
>
>>>
>>> If the handy-box is steel, then the screws holding the brass cover plate will
>>> be a sufficent ground. (if the handy-box itself is, as the NEC requires,
>>> grounded independent of the device).
>>
>> That's not the way code treats it... look at a "spec grade" device, it
>> generally is self-grounding as well, there's a brass colored clip on one
>> end of the yoke making a more positive connection with the screw holding
>> the yoke to the box. Cheaper devices won't have that but will still
>> definitely have a ground screw. The older devices don't have that nor
>> do they have a ground screw connected to the yoke. The only ground path
>> is provided by the yoke to box screws which isn't currently considered
>> sufficient. (in practice, it probably is... I'm just being pedantic here.)
>
> If by "yoke" you are referring to the bit on the device (i.e. switch body)
> that contacts the screw which holds the device into the box and is directly
> connected to the grounding post on the device, that's not sufficient by
> code to ground a non-grounded metal handybox.
Agreed. I'm thinking of the instance of a plastic single gang switch
box where the box itself is not grounded but you want to use a metal
cover plate. And the "yoke" is in fact the metal strap with the holes
in it to connect the device to the box, it has a specific definition.
> You still must, per NEC,
> attach the box directly to the ground (generally by pigtailing the grounding (not grounded)
> conductor feeding the box or alternatively by using listed conduit fittings
> with conductive, properly grounded conduit) - you cannot rely on the grounded
> device grounding the box via the screw attaching the device to the box.
Again, we're thinking of this in the opposite direction. A metal box of
any type should always be grounded, that is not only code but a best
practice. But what I am concerned about is the instance where a hot
conductor could contact the yoke or the cover plate without a path to
ground. The fuse or breaker will not trip, therefore an occupant could
take a shot up the arm if they touch the cover plate while grounded.
> A metal cover plate is sufficiently grounded by the screws (assuming they're
> conductive) attaching it to the handy-box. Look at any UL-listed metal utility
> cover at your local electrical aisle.
But the screws aren't connection the cover to the box, they're
connecting the cover to the yoke. The yoke is screwed to the box by
6-32 machine screws. Now the odd thing is that my memory says that the
NEC considers that a sufficient ground for a snap switch (assuming a
metal box) but doesn't consider it a sufficient ground for a receptacle,
unless it has the little special clip on one of the mounting screws and
is explicitly marked as "self-grounding." I'm assuming that that will
change in future revisions of the NEC. In any case, I would be sort of
hesitant to use an antique, non-grounding type switch without a
grounding pigtail securely attached to the yoke in a metal box
(duplicating the function of the grounding screw in a modern snap
switch) and under no circumstances would I install it in a plastic box
without the same, unless it was on a GFCI protected circuit. Whether an
AHJ would find either arrangement acceptable is yet another subject
entirely, and generally they tend to err on the side of caution.