Grounding / Bonding metal roof coverings

Status
Not open for further replies.

terryb

Member
Can anybody tell me whether or not metal roof coverings need grounding, bonding or both. Cannot find the answer for love nor money, would appreciate some help. Thanks, Terry B.
 

dereckbc

Moderator
Staff member
Location
Plano, TX
Good question but the answer lies in 250.4(B)(3). You will have to answer the question yourself after reading. If in doubt bond it.
 

wbalsam1

Senior Member
Location
Upper Jay, NY
Bonding of metal roofs, sidings & veneers

Bonding of metal roofs, sidings & veneers

Hi: In New York State we used to have a section in our old State code, at 1030.1(g) that spelled it right out. it said: "(g) Metal roofs, veneers and sidings on buildings shall be made electrically continuous and shall be grounded in at least two locations on the opposite sides of the building in conformity with Article 250 of reference standard RS 51-2."(1993 NEC)
A problem arose when building officials only required the metal be connected to the two ground rods driven at opposite sides, but did not realize the importance of connecting these rods to the power system grounding electrode system, thereby creating hundreds of high-resistance paths throughout their jurisdictions. When we adopted the International Family of Codes, we fell back to 250-116 in the 1999 NEC.
Now the discussion/argument is centered in the grey area contained in the phrase "likely to be energized." One side believes roofs, sidings and veneers are likely to be energized at any time by a workman's leadcord, etc., while others say only masts thru the roof or holiday lighting, etc., will actually energize the metal. This area could sure use clarity.
I do think that the 2005 NEC says it pretty well at 250.4 (A)(4) & (5)
 
Last edited:

wbalsam1

Senior Member
Location
Upper Jay, NY
Hi: George, You're saying that the egc will provide the ground-fault return path, and that bonding may be done if one so desires. True? I agree.
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Well, sorta. :)

I'm saying there is nothing to energize the siding/roofing, so there is no need to bond it. :)

250.104(C) requires structural metal to be bonded if likely to become energized. Siding and roofing are generally not structural, they are not supporting the weight of the structure.

Sheet metal screws are generally used to secure the sheets together, and these are not acceptable for bonding (250.8). If one decides to bond the siding/roofing, above and beyond code, then this isn't terribly relevant.

What are we envisioning energizing the siding/roofing?
 

infinity

Moderator
Staff member
Location
New Jersey
Occupation
Journeyman Electrician
georgestolz said:
:confused:

Roofs and siding do not require grounding or bonding.

They can be bonded or grounded if desired.


Ditto. These aren't likely to become energized. Maybe we should ground or bond metal fences.
 

wbalsam1

Senior Member
Location
Upper Jay, NY
georgestolz said:
Well, sorta. :)

I'm saying there is nothing to energize the siding/roofing, so there is no need to bond it. :)

250.104(C) requires structural metal to be bonded if likely to become energized. Siding and roofing are generally not structural, they are not supporting the weight of the structure.

What are we envisioning energizing the siding/roofing?

What about "Holiday Lighting" attached to the metal eaves, or a conduit service mast thru the metal roof...? would the egc of the lighting branch circuit and the grounded conductor in the riser satisfy the code's intent to carry the fault currents? and would the code consider these scenarios "likely to be energized" and therefore require additional bonding/grounding?
 

dereckbc

Moderator
Staff member
Location
Plano, TX
georgestolz said:
:confused:

Roofs and siding do not require grounding or bonding.

They can be bonded or grounded if desired.

George, I pretty much agree with you with exceptions of what ifs? Like the weatherhead passes throught the metal roof or overhead lines pass over a metal roof.
 

wbalsam1

Senior Member
Location
Upper Jay, NY
wbalsam1 said:
What about "Holiday Lighting" attached to the metal eaves, or a conduit service mast thru the metal roof...? would the egc of the lighting branch circuit and the grounded conductor in the riser satisfy the code's intent to carry the fault currents? and would the code consider these scenarios "likely to be energized" and therefore require additional bonding/grounding?

I'd love to hear more on this....:-?
 

infinity

Moderator
Staff member
Location
New Jersey
Occupation
Journeyman Electrician
IMO not one of the scenario's presented in this thread have any chance of being consider likely to become energized. The key word being likely. Is there an infinitesimal possibility that it may happen? Yes, but that doesn't make it likely.
 

bphgravity

Senior Member
Location
Florida
Here is one reason why bonding a metal roof may be beneficial. The quote comes from a comment made by Professor Mousa of the University of Florida when asked a question similar to that of the OP:

When lightning strikes a (dry) metal roof, it will leave a burn mark which may
develop into a hole. Also, the area surrounding the contact point will show
"blistering" because of heating effects. On the other hand, when lightning
strikes a body of water or soil, and considering that the medium is mostly
non-uniform, the current will flow in discrete radial channels. This is best
exhibited by a photo of a lightning strike to a golf course that was published
by Professor E.P. Krider in the June 1977 edition of Weatherwise, Vol. 30, No.
3, pp. 111 and the cover.

A sheet metal roof is usually supported by a frame of larger cross section
beams. The current of the stroke terminating on the water above it will tend to
flow radially as in case of strikes to deep water. However, the shallow depth
(a couple of inches) causes the current to flow instead radially along surface
of the metal. This pattern will be distorted by the attraction effect of the
frame beneath it. The metallic filaments being heated are subject to the
cooling effect of the water on the upper surface and no such cooling on the
lower surface. The difference in this "adiabatic expansion" between the top and
bottom surfaces can cause the metall to be ripped along the channels in which
the current is concentrated. This would then explain what you observed.

Abdul M. Mousa, Ph.D., P. Eng., Fellow IEEE
 
wbalsam1 said:
What about "Holiday Lighting" attached to the metal eaves, or a conduit service mast thru the metal roof...? would the egc of the lighting branch circuit and the grounded conductor in the riser satisfy the code's intent to carry the fault currents? and would the code consider these scenarios "likely to be energized" and therefore require additional bonding/grounding?


The service mast going through the roof?
If the service conductors inside of the mast, on the line side of the service disconnect were to fault to the metallic mast raceway, what overcurrent device on the premise is going to open, and stop the current flow from the fault on the line side?...it will continue faulting until the primary opens (most unlikely) or it will burn clear. Therefore the bonding of the metallic roof will not help here.

Holiday lighting?
Holiday lighting itself does not have an equipment ground installed within its cable. The circuit supplying it should. If we were to bond the roof or siding, will it open the circuit breaker...it may. One important requirement of holiday lighting, is it is limited as to how long it is to be installed. See 410.110, which refers one to 590.3(B) - essentially 90 days & UL requirements as well.
 

LawnGuyLandSparky

Senior Member
And another requirement of holiday lighting is a 3a fuse in the plug.

MVC-433S-1.jpg

MVC-432S-1.jpg

MVC-430S-1.jpg
 

RUWIREDRITE

Senior Member
Bonding of metals

Bonding of metals

Hi all,
I would love seeing a change in the new nec code to bond any metal structures in residential that exceed a certain size, a good example is I went to a home once with an unrelated electrical issue,during my investigation I had to remove some tiles from a drop ceiling grid. During my visit in this ceiling my neck came in contact with the grid , at the same moment I grabbed a plumbing pipe.I got knocked clear off the ladder, later to find out the builder ran a sheetrock screw thru a piece of romex when he install the L channel on the wall. It energized the entire drop ceiling grid and the grid was insulated from any grounded serfaces till i grabbed the pipe.Thank god it wasnt a 30 foot ceiling,lol.
 

bphgravity

Senior Member
Location
Florida
RUWIREDRITE said:
Hi all,
I would love seeing a change in the new nec code to bond any metal structures in residential that exceed a certain size...

Ryan Jackson sent in a nice proposal for just this and it was soundly rejected by CMP 5. See proposal 5-228 of the ROP.

Florida has amended the Florida Building code to require bonding of metal studs in any building. See E3303 of the FRC, and 2704 of the FBC.
 
bphgravity said:
Ryan Jackson sent in a nice proposal for just this and it was soundly rejected by CMP 5. See proposal 5-228 of the ROP.

Florida has amended the Florida Building code to require bonding of metal studs in any building. See E3303 of the FRC, and 2704 of the FBC.


Bryan

Has this truely made any difference?
Does it take the pressure off of installers to think of possibly doing the work correctly? Is not the installation already bonded using metallic boxes? Questions, questions, questions!!!

I am always curious about changes such as this...what about all of the existing installations, of which they are too numerous to really think about. Those existing installations will most likely be around a very long time. Are they any kind of problem to think of, such as the new installations now are???


I am not picking on you Bryan, I just have questions...with your knowledge or someone else here, maybe I can quench my thirst. ;)
 

bphgravity

Senior Member
Location
Florida
Pierre,

I actualy submitted a comment against the Florida amendment. I will post the original proposal and my comment on Monday when I get back to my office.

The proposal was made after the death of an appliance installer when he came in contact with an energized dryer vent duct box in the wall of a new home.
 

wbalsam1

Senior Member
Location
Upper Jay, NY
Conductive roofing or siding shall be bonded in NY

Conductive roofing or siding shall be bonded in NY

:) In New York State, we have included enhancements to the stand-alone residential code (chapters 33-42) which are based on the NEC 2002. One such enhancement has to do with bonding of metal roofs and sidings. I've underlined our enhancement.
At Section E3509 BONDING we authored this:
E3509.1 General. Bonding shall be provided where necessary to ensure electrical continuity and the capacity to conduct safely any fault current likely to be imposed. Conductive roofing or siding shall be bonded when electrical equipment is mounted on the roofing or siding.

Of course this is not prescriptive, in the sense that we do not say "how" to do it. :) But this particular text is in a code yet to be implemented, perhaps by August, I'm told.
 
Status
Not open for further replies.
Top