Announcement

Collapse
No announcement yet.

Here's one for ya...

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Here's one for ya...

    Say I have a structure. This structure has Pv modules for the Roof - as in modules laid right on trusses. Say I seal the seams, use gutters underneath the seams, shingle the panels, have mice hold up umbrellas over the seams, some combination thereof, whatever. The structure has floors and walls. Is it now a building? Does it need a "real" roof to be considered a building? If its a building, it needs RSS. What is inside vs outside the building for consideration of controlled conductors? IF its a building, I dont see any way to comply with RSS.
    Ethan Brush - East West Electric. NY, WA. MA

    "You can't generalize"

    #2
    Click image for larger version

Name:	20181222_WBP001_1.jpg
Views:	1
Size:	73.9 KB
ID:	2380226
    Sort of like this photo from an article in the Economist (British magazine)? Kind of hard to understand code compliance if this photo is of a US installation. Seems plenty of gap to allow water to drain through. The article title is "Rooftop solar remains marginal in America". In Europe there are PV modules designed to be integrated into a building and be the primary roof.

    Comment


      #3
      I'll bite. AC modules do not have any controlled conductors (no PV source circuits or PV output circuits) and would therefore not require rapid shutdown.

      Comment


        #4
        Bill,

        Great photo! Ive been looking for an example of this and cant find anyone doing it. I am doing it at my house (not on my house but on an outbuilding). My inspector said it doesnt have a roof so its not a building, but RSS seems to be a real killer for something like this in more strict (fascist) areas. I'll provide more information on my system and how I am dealing with mounting and the seams coming up.

        PWD,

        I thought about micros. That may work. We are under 2014 still here. When looking at 690.12(1) perhaps one could get the AHJ to go with the "10 feet from the array" distance, since there isnt really a clear building envelope separating the inside from the outside.
        Ethan Brush - East West Electric. NY, WA. MA

        "You can't generalize"

        Comment


          #5
          Originally posted by electrofelon View Post

          I thought about micros. That may work. We are under 2014 still here. When looking at 690.12(1) perhaps one could get the AHJ to go with the "10 feet from the array" distance, since there isnt really a clear building envelope separating the inside from the outside.
          Maybe I am misinterpreting what you said, but AC modules (microinverters) are inherently RSS compliant no matter what the code cycle.

          Comment


            #6
            Originally posted by ggunn View Post
            Maybe I am misinterpreting what you said, but AC modules (microinverters) are inherently RSS compliant no matter what the code cycle.
            I should have put a paragraph in there. The second part was assuming micros/ac modules are not being used.
            Ethan Brush - East West Electric. NY, WA. MA

            "You can't generalize"

            Comment


              #7
              Sounds like a building to me.

              The 'array boundary' under 2017 is one foot from the array in any direction. To me that includes 1ft on the bottom side, whether that is inside or outside the building. Optimizers or micros or whatever basic RSS function MLPE people are using on most buildings would still be compliant in your scenario.

              Comment


                #8
                RSS doesn't worry me as much in this scenario as wiring method requirements for PV source circuits. I forget if the exception that applies to underneath the array would apply here.

                Comment


                  #9
                  Originally posted by jaggedben View Post
                  Sounds like a building to me.

                  The 'array boundary' under 2017 is one foot from the array in any direction. To me that includes 1ft on the bottom side, whether that is inside or outside the building. Optimizers or micros or whatever basic RSS function MLPE people are using on most buildings would still be compliant in your scenario.
                  Originally posted by jaggedben View Post
                  RSS doesn't worry me as much in this scenario as wiring method requirements for PV source circuits. I forget if the exception that applies to underneath the array would apply here.
                  I agree its a little more straightforward in 2017. Under 2014, the way I read it, it doesnt seem doable with string inverters. Good point about the wiring method, thats a concern too.
                  Ethan Brush - East West Electric. NY, WA. MA

                  "You can't generalize"

                  Comment


                    #10
                    Originally posted by electrofelon View Post
                    I agree its a little more straightforward in 2017. Under 2014, the way I read it, it doesnt seem doable with string inverters. Good point about the wiring method, thats a concern too.
                    Under the 2014 NEC we put string inverters on the roof 10 feet or less from the arrays and we spaced subarrays 20 feet or less apart. Until last week under the 2017, 10 feet and 20 feet was changed to 1 foot and 2 feet.

                    Comment


                      #11
                      Originally posted by electrofelon View Post
                      I agree its a little more straightforward in 2017. Under 2014, the way I read it, it doesnt seem doable with string inverters. Good point about the wiring method, thats a concern too.
                      I reread 2014 and saw the 'PV conductors of more than 5ft inside a building', which would almost certainly include your module to module conductors. So I agree that under that you can't do it with string inverters unless you can win the 'inside the building' part of the argument. Or unless you could convince your AHJ that the conductor length is only counted from the module j-box to the connector!

                      Maybe if you employ the same methods that are sometimes employed to 'guard' conductors on readily accessible ground mounts, then you could convince an AHJ that the conductors are not 'inside' the building. i.e. since they are contained inside the 'roof'. i.e. the way the roofing paper between shingles and plywood isn't 'inside' the building. If I were an AHJ I would probably accept that, but I'm not one.

                      Comment


                        #12
                        With regard to my last paragraph of the previous post, and the 2017 code, see 690.12(B)(3). If you can build the array with frameless modules and 'no exposed wiring methods' then you could still use a string.

                        Comment

                        Working...
                        X