[RE-wrenches] The perfect solar ready roof

Fri Oct 22 11:22:32 PDT 2010

Roof construction knowledge and roofing experience is essential if you are a solar designer and installer. You or people you work with must with have roof experience. If you do not have an experienced roofer in your organization, then you must work closely with an experienced roofing subcontractor (unless you like getting angry phone calls when it is raining).

Framing and roofing a standing seam metal roof is not difficult. There is no need to guess or to ask inexperienced people what they think should be done. Good suppliers will provide specifications, instructions, and drawings and put you in touch with experts who can answer your questions.

In 1996, I did the Unisolar standing seam roof at the University of California Irvine campus. See http://www.hamcontact.com/unisolar/roofapps.html All we had was to start with was a 2nd floor deck and a contract with Southern California Edison (SCE) and the DOE UPVG (Utility Photovoltaic Group now called SEPA) who cost-shared the 5 kW new roof BIPV system. Unisolar provided standing seam metal roofing installation instructions but knew very little about roofing. I talked with McElroy, the manufacturer of the roofing Unisolar used for their SSR product. I also visited metal roofing jobsites during construction to see if anything new had been developed since I last did a metal roof a few years before. Then I designed the framing, building attachments, and the PV system and got SCE and Campus approval. I was concerned about my attachments to the roof deck knee-wall so I asked for the building structural drawings and really lucked out. The engineer who had designed the building still worked at his nearby office part-time even though he was 85 years old. (This old-timer still came to his office in a suit and tie to share his over 50 years of knowledge, experience, and wisdom. I love working in the construction industry.) He and I spent 15 minutes reviewing my design and he gave it his blessing.

When I designed the solar roof, the University's Combustion Engineering Lab Director said that he might want to someday enclose the roof deck so we put in metal diagonal straps under the roof panels to brace the roof. The engineer said this was a good idea because it strengthened the roof in case of earthquakes.

The construction went smoothly. We fastened the ledger and post attachments with lag bolts and attached the posts, beams and purlins. The SSR panels were laminated at the Unisolar factory and shipped from Troy, Michigan (a waste of time and money). When we started fastening the standing seam panels, I had to ask the University lab engineers to turn off the experimental natural gas fuel cell they were testing on the deck because the rising heat was cooking us on the roof like grilled hotdogs. We fastened clips 6 inches from panel ends and also every 12 inches using self-taping stainless steel screws (over-kill but it went fast). The slowest work was lapping the 12-feet panels over the 18-feet panels because each panel's standing seam had to be notched with a sidecutter and fit into place. The electrical work was conventional. The homerun in conduit went into the building to a Trace SWODE SW5548 that back-feeds to a 120 VAC breaker in a subpanel.

A few years ago, I was on campus so I dropped by to inspect the system. The deck was enclosed and the space turned into offices and the Trace inverter was still working like new.

Joel Davidson

----- Original Message ----- 
  From: North Texas Renewable Energy Inc 
  To: RE-wrenches 
  Sent: Friday, October 22, 2010 7:47 AM
  Subject: Re: [RE-wrenches] The perfect solar ready roof

  Phil hits the nail on the head, it sounds like we've both been down this road. Problem is it's not always as simple as asking the roofers to do their job your way when you are not signing their paycheck.
  I contracted a metal roof PV installation a few years back but only got the job on the condition, by the very fastidious homeowner, that I could verify that the installation won't cause leaks or void the warranty on the <2 year old roof. He gave me the roofers name and the roofing product and mfgr. I researched the manufacturers installation requirements for our inland climate. Then I called the roofer and asked if the 16" wide panels were installed per the mfgrs specs. 
  He proceeded to go ballistic, telling me that if I put one module on 'his' roof he would void the warranty. Even after sending him links to the S5 engineering test results he refused to even talk to me. Finally the customer had to have a long talk with the guy before he was convinced his roof had been installed correctly. 
  Bottom line you cannot ever be certain about the quality of the installation by someone else.  Besides, the homeowners insurance company will have the last word on whether a PV mounted roof, found across the street after a windstorm, was installed right or not. 
  CYA
  Jim Duncan
    -----Original Message-----
    From: re-wrenches-bounces at lists.re-wrenches.org [mailto:re-wrenches-bounces at lists.re-wrenches.org]On Behalf Of Phil Undercuffler
    Sent: Thursday, October 21, 2010 10:29 PM
    To: RE-wrenches
    Subject: Re: [RE-wrenches] The perfect solar ready roof

    There is a big difference between standing seam metal roofs and surface screwed metal roofs.  With surface screwed (aka ProPanel or Ag Panel), screws penetrate through the roof skin and rely upon a neoprene gasket under a cupped washer to provide the weather-tight seal.  It makes a pretty good seal and I've used it on my own home, but I don't think I'd use it in Montana and I sure wouldn't try to attach PV over it.  The days of lagging L feet through the roof deck are thankfully waning, and there is almost no practical way to use a flashed method of mounting with surface screwed metal. 

    With standing seam roofs, the metal is formed into long U shaped pans. As each pan is installed, the roofer nails L shaped "clips" to the deck with one side of the clip butted up against the latest pan.  The next pan is butted up against the first, which captures the clip between the two.  The upward facing legs of the U (and clip) are then crimped and folded over, locking them together and forming a watertight seal.  I had a standing seam roof on my home in Cincinnati -- eighty years old and still going strong.  

    The biggest challenge with attaching PV to standing seam roofs is not how well the modules are attached to the skin, but how well the skin is attached to the structure.  Read the archives, but in the end this really comes down to how close the roofer installed the clips.  This is where the 4" that Andrew mentioned comes into play (seems a little excessive to me, but what the heck, I'm not a roofing dude and this isn't something you want to do twice). The dream scenario part comes in when you realize that YOU get to drive the bus on how closely the clips are installed, as opposed to being presented with a roof where you have no idea how many clips were used, what the spacing is, or even whether nails or screws or bubble gum were used to connect them to the deck.  The folks that make the S-5 clamp know metal roofs, and they can help you determine what spacing works for your application.  Get that into the contract, and make sure someone is on site providing oversight when the roof is installed.

    As far as some of the other options which were presented -- yes, lagging into engineered I-beams is probably not a good idea without checking with the manufacturer, but adding a second layer of ply doesn't suddenly make a structural base for lag screws.  Lag screw pull-out resistance is provided by inches of thread embedded into solid wood, and shiners (fasteners that poke through the deck into the attic space) won't give any real strength. If it's really 1/2" ply, adding a layer of 3/4 ply would give you 1-1/4" of "wood" -- do the math and see if that's enough for your climate. Adding 2x blocking would be an option if you have access to the attic, but you need to make sure the loads transfer to the rafters (I-beams), rather than concentrate on the deck.  Nailing the blocking would be fine -- after all, that's how houses are built.  Lagging the blocking as suggested by someone earlier, however, will not only be insanely difficult but risk splitting the top chord of the beam.  If you split that, I would immediately stop what you're doing and consult with the I-beam manufacturer.  Big liability moment there.

    But I'd avoid all that lags, nails, plywood and blocking entirely, put on a standing seam metal roof with adequate clips and install the PV with S-5 clamps.

    My .02

    Phil Undercuffler 

    On Thu, Oct 21, 2010 at 11:52 AM, benn kilburn <benn at daystarsolar.ca> wrote:

      andrew, 
      a few comments added to your last email...

      You Wrote.... 

      If you ask me this is a dream scenario.  The scary thing about S-5!s is that you rarely know how often the roofing panels are attached to the decking and how well the decking is attached to the framing. i'm not overly familiar with standing seam metal roofs (and i know they are not all created equal) but aren't the screws holding the metal roof panel to the decking visible at the bottom, at the top under the vent cap and anywhere along the length of said panel?  i agree about the 'unknown' attachments btwn the decking and the framing   In this case you can direct the roofer to fasten the roof as often as you want.  I had a PE do an analysis for a sure-fire acceptable attachment method for a standing seam roof given: 90 mph wind zone, 18" wide standing seam roofing panels, flush-mounted PV array (modules to rail to S-5!s, no tilt legs), and his result was that if the roofing panels are fastened every 4" along each seam you are in the clear.  No doubt, that seems like quite a few fasteners.  He did not address the decking-to-framing attachment, So worst case scenario, you get a 90+ mph wind that takes the array, the metal roof and the decking for a ride to the ground, (HAS ANYONE SEEN OR HAD THIS HAPPEN?) when it is determined that the decking wasn't properly attached to the framing, who do ya call?  PV installer, PE who stamped it, roofer?  I'm not looking for a place to point fingers, I'm just looking for some insight on this so i understand our responsibilities a bit better.  I realize that it would be onerous to confirm the number and placement of decking to framing attachments)  but while the roof is off you could add as many fasteners as you want.  If the structure of the roof is in question I would definitely have a PE look at it and stamp the plans before proceeding, but once you have everything under the roof robust and approved, you are primed and ready for a worry-free, penetration-free S-5!-to-standing-seam array install.

      In another email string someone mentioned issues with S-5!s slipping down the roof due to snow which is why I recommend attaching at every seam and breaking out that torque-wrench when installing.  

      Good luck!

      ...end of your msg

      in response to Kris' suggestion, don't the 2x6's need to be toe-nailed into the 'rafters' on either side? or in this case toe-nailed into the beams? which probably wouldn't be much different than the 'inadvisable' lag bolting into them.

      cheers,
      benn
      DayStar Renewable Energy Inc.  
      benn at daystarsolar.ca
      780-906-7807 
      HAVE A SUNNY DAY 

--------------------------------------------------------------------------

      From: atruitt at gmail.com
      Date: Thu, 21 Oct 2010 11:30:40 -0600
      To: re-wrenches at lists.re-wrenches.org 

      Subject: Re: [RE-wrenches] The perfect solar ready roof

      If you ask me this is a dream scenario.  The scary thing about S-5!s is that you rarely know how often the roofing panels are attached to the decking and how well the decking is attached to the framing.  In this case you can direct the roofer to fasten the roof as often as you want.  I had a PE do an analysis for a sure-fire acceptable attachment method for a standing seam roof given: 90 mph wind zone, 18" wide standing seam roofing panels, flush-mounted PV array (modules to rail to S-5!s, no tilt legs), and his result was that if the roofing panels are fastened every 4" along each seam you are in the clear.  He did not address the decking-to-framing attachment, but while the roof is off you could add as many fasteners as you want.  If the structure of the roof is in question I would definitely have a PE look at it and stamp the plans before proceeding, but once you have everything under the roof robust and approved, you are primed and ready for a worry-free, penetration-free S-5!-to-standing-seam array install.
      In another email string someone mentioned issues with S-5!s slipping down the roof due to snow which is why I recommend attaching at every seam and breaking out that torque-wrench when installing.  

      Good luck!

      Andrew Truitt
      NABCEP Certified PV Installer™ (ID# 032407-66)
      Truitt Renewable Energy Consulting
      (202) 486-7507
      http://www.linkedin.com/pub/andrew-truitt/8/622/713

      "Don't get me wrong: I love nuclear energy! It's just that I prefer fusion to fission. And it just so happens that there's an enormous fusion reactor safely banked a few million miles from us. It delivers more than we could ever use in just about 8 minutes. And it's wireless!"

      ~William McDonough

      On Wed, Oct 20, 2010 at 2:32 PM, Kristopher Schmid <solman at legacysolar.com> wrote:

        What about screwing in double 2x6s flush to the roof deck between the beams where your feet will attach and lag bolting into that?  Definitely check with the beam manufacturer first, though.

        Kris
        Legacy Solar
        864 Clam Falls Trail
        Frederic, WI 54837
        715-653-4295
        solman at legacysolar.com
        www.legacysolar.com 

          -----Original Message-----
          From: re-wrenches-bounces at lists.re-wrenches.org [mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Scott McCalmont
          Sent: Wednesday, October 20, 2010 12:00 AM
          To: RE-wrenches
          Subject: Re: [RE-wrenches] The perfect solar ready roof
          In general, you shouldn't drill or cut the flanges on engineered wood beams. I think that eliminates lag screws into the rafters. They probably wouldn't have the same pull-out strength as a lag screw into a conventional rafter, either. 

          Scott

          On Oct 19, 2010, at 7:58 PM, Chris Daum wrote:

        Dear Wrenches:
        I have a composite (shingle) roof at hand, and the owner wants to upgrade it to a metal roof and install a 5kw+ array on it.   The rafters are those (sort of) particle board I-beams covered with 1/2" plywood (and shingles).  What's the best metal roofing you could suggest--and would you beef up the wood to lag into?
        Thanks for all your input.
        Chris Daum
        Oasis Montana Inc.
        406-777-4309
        406-777-0830 fax
        _______________________________________________
        List sponsored by Home Power magazine

        List Address: RE-wrenches at lists.re-wrenches.org

        Options & settings:
        http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

        List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

        List rules & etiquette:
        www.re-wrenches.org/etiquette.htm

        Check out participant bios:
        www.members.re-wrenches.org

      _______________________________________________
      List sponsored by Home Power magazine

      List Address: RE-wrenches at lists.re-wrenches.org

      Options & settings:
      http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

      List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

      List rules & etiquette:
      www.re-wrenches.org/etiquette.htm

      Check out participant bios:
      www.members.re-wrenches.org

      -- 

      _______________________________________________ List sponsored by Home Power magazine List Address: RE-wrenches at lists.re-wrenches.org Options & settings: http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org List rules & etiquette: www.re-wrenches.org/etiquette.htm Check out participant bios: www.members.re-wrenches.org

      _______________________________________________
      List sponsored by Home Power magazine

      List Address: RE-wrenches at lists.re-wrenches.org

      Options & settings:
      http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

      List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

      List rules & etiquette:
      www.re-wrenches.org/etiquette.htm

      Check out participant bios:
      www.members.re-wrenches.org

------------------------------------------------------------------------------

  _______________________________________________
  List sponsored by Home Power magazine

  List Address: RE-wrenches at lists.re-wrenches.org

  Options & settings:
  http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

  List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

  List rules & etiquette:
  www.re-wrenches.org/etiquette.htm

  Check out participant bios:
  www.members.re-wrenches.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20101022/0f60a2dc/attachment-0004.html>