[RE-wrenches] Attaching Array To or Through Ply
Jason Szumlanski
jason at floridasolardesigngroup.com
Fri Mar 31 13:15:38 PDT 2023
Hi Ray,
I'm happy to have some confirmation from another major storm's impact. My
father has a home in Dominica, which was hit by Maria with 165 mph winds
before it moved on to PR. He had a mixed bag of results with solar panels
(forget about the pole mounts I installed there over 20 years ago!)
I was just reminded by your email about something I plan to try. For panels
that are pitched on flat roofs, I am going to double up and install BOTH
IronRidge CAMO and UFOs. With proper planning, it would be pretty easy to
install with UFOs first and then go back and snap in all of the CAMOs
without much labor cost and an acceptable material cost for the added peace
of mind.
Jason Szumlanski
Principal Solar Designer | Florida Solar Design Group
NABCEP Certified Solar Professional (PVIP)
Florida State Certified Solar Contractor CVC56956
On Thu, Mar 30, 2023 at 9:45 PM Ray Walters via RE-wrenches <
re-wrenches at lists.re-wrenches.org> wrote:
> Hi Jason;
>
> I was like you: thinking good ol' bottom mounting with SS hardware was the
> best. All of us old wrenches were wrong.
>
> The top down mounting system is much stronger. I've only had a handful of
> failures over 25 years, and all were bottom mounted. The module completely
> ripped off the rack, leaving the stainless hardware, washers, etc with a
> thin sliver of module frame between. For repairs, I come back and install
> 1/2" SS angle on the inside of the module frame to spread the load more.
> AND....on new installs, no more bottom mounting. If you compare today's
> modules with old 12v models from 20 years ago, you'll see: much thinner
> metal on the bottom frames now.
>
> Meanwhile I saw several systems survive Hurricane Maria in Puerto Rico:
> all top down mounting. You are correct that the failures come from the
> modules themselves letting go, not the racking. Either the wind itself is
> so strong as to blow the glass, then the frame looses its structure, or
> flying debris breaks the glass. There were some MW PV fields that got tore
> up pretty bad. So no, ground mounts definitely can be destroyed. They had
> tornadoes moving inside of the hurricane, and you could see it in the
> damage, 50' wide strips of total carnage with undamaged modules a few feet
> away. Like you said, The storm has a mind of its own.
>
> As far as mid clamp T bolt failures, I can confirm that probably a
> majority are not installed correctly, especially Unirac, which are terrible
> to get the T lined up right. I've done numerous inspections world wide,
> and a signifcant % of racking I looked at was NOT installed correctly.
> 1) The Ts not squarely lined up with the rail,
> 2) massive over torqueing, to the point the bolt can't be reversed
> 3) under torqueing, to the point the modules had slipped down a little.
>
> Most of these problems come from installing with an impact driver. Impact
> drivers are a great tool (so is a hammer), but you got to know when and
> where to Not use them, too. They're great for installing the lags and L
> feet, but then get that tool off the roof.
>
> Use an electric screw driver, socket wrench, or *drill with a clutch* set
> low to snug up the clamps, and then final torque with a torque wrench.
> Period.
>
> Ray Walters
> Remote Solar
> Former NABCEP 2004-2016
>
>
> On 3/30/2023 4:34 PM, Jason Szumlanski via RE-wrenches wrote:
>
> That would be great if it were possible! I'm not sure how you bottom mount
> a flush mount array. For a brief period many years ago, I was building
> solar trailers. I would bottom mount 4-6 panel arrays on rails in my
> warehouse and then hoist them up and through-bolt the whole array on the
> top of a trailer. I wasn't about to allow a trailer going 80 mph on the
> interstate to have panels mounted on the roof with mid-clamps! I had no
> idea what I was doing, but I had enough sense to know that would have been
> a bad idea!
>
> I did provide feedback to IronRidge regarding a handful of failures from
> Hurricane Ian, but most of the issues we saw were most likely related to
> catastrophic module failure. After all, they are only tested to 5600 Pa
> uplift for even the best options we have on the market and 2400 Pa for some
> (which I refuse to use). The mid-clamp T-bolts tearing out of the rail are
> slightly concerning, but this was pretty rare, and honestly it could be
> related to installation torque issues, misalignment, or coupled with module
> failure. There was no widespread or definitive reason for rail failures at
> the clamp locations.
>
> I would suggest that FEMA's recommendation is overkill and not based on
> much science, but conclusions based on anecdotal information without
> statistical data. It should be scrutinized. I think it would be a good idea
> to consider bottom mount for ground racks and tilt mounts that allow it,
> but it's just not practical or possible in the vast majority of residential
> installation cases. On that note, I am not aware of any ground mount
> failures around here from Hurricane Ian except for submerged arrays that
> were washed away from storm surge (Yikes!). And the handful of failures
> that we observed were a drop in the bucket relative to the installed
> numbers here.
>
> One other thing. Wind direction, upwind obstructions/windbreaks, and luck
> have a huge amount to do with failures (of both PV and roofs themselves).
> These storms pick winners and losers. You will have ten houses in a row
> with pool enclosures mangled, and one in the middle that is unscathed. It's
> crazy to see. We have lots of gated communities with houses close together.
> When wind accelerates between houses, it can topple air conditioning units
> and pool equipment. If you look hard enough at where the wind was coming
> from and the surrounding area, you can really see how there are so many
> factors that come into play.
>
> Jason Szumlanski
> Principal Solar Designer | Florida Solar Design Group
> NABCEP Certified Solar Professional (PVIP)
> Florida State Certified Solar Contractor CVC56956
>
>
> On Thu, Mar 30, 2023 at 7:35 AM Christopher Warfel via RE-wrenches <
> re-wrenches at lists.re-wrenches.org> wrote:
>
>> If you look at FEMA's design guide for solar installations in FL and the
>> Caribbean, the recommended module to rail attachment method is back to
>> using the attachment holes in the solar module. Chris
>> On 3/30/2023 5:56 AM, Jason Szumlanski via RE-wrenches wrote:
>>
>> This thread has morphed into more than it started as, and for good
>> reason. I want to provide some practical/anecdotal information, having just
>> gone through arguably the most catastrophic wind event in Florida's history
>> (Ian) since the boom in solar energy started, and another major wind event
>> (Irrma) just 5 years ago.
>>
>> First, Aside from the Sunmodo and Quickbolt decking-only products, I
>> would check out IronRidge's new entrant into the market, the IronRidge HUG.
>> If you can't find info on it, ask your distributor or IronRidge sales rep.
>> They have data on truss attachments and also missed truss installation
>> procedures. It's a unique approach to a dual-purpose product, and they did
>> a stellar job on the engineering documentation in my opinion.
>>
>> Ok, now let's get to my most important point. Due to Hurricane Ian, the
>> number of roofing PV attachments that I have witnessed that failed due to
>> fastener pull-out (mine or competitors):
>>
>> ZERO
>>
>> The only building where we had any mounts fail was on a flat roof with
>> pitched panels and Anchor Products mounts on TPO membrane, but the mounts
>> themselves did not cause the failure. The roofing system itself failed,
>> causing a small section of the PV system to fail. But even in that case,
>> the following applies...
>>
>> The weak point in a well-designed and installed system is not the
>> fastener or flashing system. The module to rail connection is where we saw
>> failures. These failures fell into a few categories:
>>
>> - Windborne debris struck panel, panel frame failed, panel popped out
>> of mid-clamps.
>> - Catastrophic wind forces popped panels out of mid-clamps (a good
>> percentage of panels found INTACT and still functional on the ground!) I
>> suspect the panels became covex in the wind, bending frames inward.
>> - Windborne debris struck mounting system components, panel
>> dislodged, often still on the roof suspended by DC leads.
>> - Mid-clamp t-bolt tore out of aluminum rail channel (IronRidge UFO,
>> Unirac SM).
>> - Mid-clamp sheared off (Quick Mount QRail).
>> - Unexplained module detachment failures.
>>
>> On 9/28/22, while I stayed up all night bracing myself against my front
>> door that I thought was about to fail, I was imagining how many roof leaks
>> my clients were about to endure, and wondered about the efficacy of my
>> business going forward. Those fears never materialized. Aside from a
>> handful of minor panel dislodgements, there was no panic following the
>> storm (with respect to solar panels). The bigger problem became all of the
>> people needing to remove panels for roof replacements, but PV panels
>> largely protected roofs in the areas where they were installed. Sadly, the
>> rest of the roof often did not fare as well.
>>
>> Anyway, back to the decking attachments. I have been skeptical of
>> non-flashed products for comp shingle roofs for a long time. My thinking is
>> coming around, particularly with the HUG (I trust IronRidge's testing
>> regime). And sealants have come so far. This method will remain up for
>> debate probably for a long time. Around here, I am pretty certain these
>> products will outlast the shingles they are placed upon. We only get 15
>> years out of most shingle roofs around here.
>>
>> About the pull-out fears... Mine are gone. We have done many flat roofs
>> with long fasteners through steel decking or wood decking. These screws are
>> usually something like #15 XHD screws in lengths from 5 - 12". Not a single
>> failure. We have also used Quick Mount QBase Low-Slope bases on pitched
>> tile roofs that were only screwed into decking with 4 fasteners each (due
>> to horizontal truss transitions) in some cases. Zero failures. But the most
>> relevant attachments I can think of that are germane to this discussion are
>> the many thousands of S-5 SolarFoot that we have screwed into decking on 5V
>> metal roofs around here. These have four screws per attachment, and S-5
>> load tests show something like 240 lbs of pull-out strength in OSB
>> (adjusted for safety factor). When engineered for our wind loads, we
>> usually get anywhere from 36-48 inch attachment spacing, sometimes 24
>> inches in certain roof zones. Again, not a single failure.
>>
>> I have more solar installations on Sanibel Island and Fort Myers Beach
>> (Hurricane Ian Ground zero) than anyone. Many of those have decking-only
>> attachments, S-5 clamps, flat roofs, or other attachments other than
>> trusses. I can tell you unequivocally that I trust decking-only attachments
>> from a pull-out strength standpoint. When properly engineered, with
>> cautious attachment spacing, these mounts work in both OSB and plywood.
>> Because of the inconsistencies in OSB, we always err on the side of
>> caution, if not in the engineering, then in the installation, by installing
>> more attachments than prescribed. But the evidence is clear. It works.
>>
>>
>> Caveat to the above: I have zero experience with snow or seismic, and no
>> experience on roofs exceeding 8:12 pitch, and few above 6:12.
>>
>> And one more shout-out to S-5 clamps on standing seam roofs. Aside from
>> one minor failure of the roof metal itself, not an S-5 failure, we had zero
>> failures of S-5 clamps attachments to report.
>>
>> I hope this anecdotal information helps and sets some fears aside. Please
>> reach out to me off-list if you want any specifics or details about our
>> experience with catastrophic wind events.
>>
>> Sincerely,
>>
>> Jason Szumlanski
>> Principal Solar Designer | Florida Solar Design Group
>> NABCEP Certified Solar Professional (PVIP)
>> Florida State Certified Solar Contractor CVC56956
>>
>>
>> On Thu, Mar 30, 2023 at 2:47 AM William Miller via RE-wrenches <
>> re-wrenches at lists.re-wrenches.org> wrote:
>>
>>> Friends:
>>>
>>>
>>>
>>> Thanks for bringing up this scenario.
>>>
>>>
>>>
>>> I have two problems with using a product such as the easy-feet or other
>>> deck-fastened brackets, particularly on a rigid foam-above-sheeting roof:
>>>
>>>
>>>
>>> 1. I worry about compression of the foam material over time. This
>>> could leave a void under the bracket which could lead to leaks or wobbly
>>> brackets.
>>>
>>> 2. I don’t believe in chemical solutions for sloped-roof
>>> rain-proofing. By that I mean the use of caulks, gaskets, etc. Since
>>> roofs were first thatched, the overriding wisdom is that gravity is the
>>> only way to reliably shed water. Overlapping, seamless material is the
>>> only method to use. Caulks degrade, roof surfaces become powdery, and the
>>> rafter is often under an architectural feature or seam in the shingles.
>>>
>>> I researched the PLP EZ foot back when the company was DPW. The
>>> fasteners provided were not rated for the application, according to the
>>> fastener manufacturer. If you can match the fasteners to the decking and
>>> the forces, then maybe you have a start to a mounting solution.
>>>
>>>
>>>
>>> Whatever product you use, I would install it on a section of flashing
>>> metal, lapped under the next course up. The flashing gives a flat,
>>> seamless surface to caulk or gasket to. The thicker and wider the
>>> flashing, the more you distribute the downward force
>>>
>>> applied to the bracket by weight and fasteners. Any voids under the
>>> bracket will have an overlapped flashing above it.
>>>
>>>
>>>
>>> I hope these musings help you find a solution.
>>>
>>>
>>>
>>> William Miller
>>>
>>>
>>>
>>> PS: I have encountered this roof configuration a few times on flat, built-up
>>> roofing (BUR <https://homeinspectioninsider.com/built-up-roofing/>).
>>> We have dealt with it by cutting through the foam, installing blocking on
>>> top of the sheeting and having a roofer feather the blocks into the roofing
>>> with cant strips
>>> <https://www.blueridgefiberboard.com/cant-strip-tapered-edge-smoothes-roof-drainage-slope/>.
>>> A pitched roof is a different situation, however.
>>>
>>>
>>>
>>> Wm
>>>
>>>
>>>
>>> Miller Solar
>>>
>>> 17395 Oak Road, Atascadero, CA 93422
>>>
>>> 805-438-5600
>>>
>>> www.millersolar.com
>>>
>>> CA Lic. 773985
>>>
>>>
>>>
>>>
>>>
>>> *From:* RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org] *On
>>> Behalf Of *August Goers via RE-wrenches
>>> *Sent:* Wednesday, March 29, 2023 11:47 AM
>>> *To:* RE-wrenches
>>> *Cc:* August Goers
>>> *Subject:* Re: [RE-wrenches] Attaching Array To or Through Ply
>>>
>>>
>>>
>>> We've found that in the Bay Area that deck-mount solutions typically
>>> pencil out structurally (via a structural professional engineer) as long as
>>> the deck is 1/2" plywood or thicker, assuming we have all the info on how
>>> the roof is constructed. The big assumption is that we can get all that
>>> roof construction info, which can be difficult for existing structures.
>>>
>>>
>>>
>>> We haven't taken the plunge yet on the flahingless deck mount products
>>> like the Sunmodo Nanomount or Unirac Flashloc Duo. I do think that they
>>> offer several significant advantages including not needing to find rafters,
>>> thus virtually eliminating missed pilot holes, and not disturbing the comp
>>> shingle by eliminating prying up the courses to insert the flashing.
>>>
>>>
>>>
>>> August
>>>
>>> Luminalt
>>>
>>>
>>>
>>> On Wed, Mar 29, 2023 at 11:34 AM Solar Energy Solutions via RE-wrenches <
>>> re-wrenches at lists.re-wrenches.org> wrote:
>>>
>>> We like Spider Rax… with the flashing!
>>>
>>>
>>>
>>> *Spider-Rax PV Solar Mounting <https://spiderrax.com/>*
>>>
>>> *spiderrax.com <https://spiderrax.com/>*
>>>
>>> *Error! Filename not specified.* <https://spiderrax.com/>
>>>
>>>
>>>
>>> *Andrew Koyaanisqatsi*
>>>
>>> President
>>>
>>>
>>> *Solar Energy Solutions, Inc. The BRIGHT CHOICE*
>>>
>>>
>>>
>>> *Since 1987, helping you and your *
>>>
>>> *Portland neighbors move towards an environmentally sustainable future.*
>>>
>>>
>>>
>>>
>>> *503-238-4502 <503-238-4502> www.SolarEnergyOregon.com
>>> <http://www.solarenergyoregon.com/>*
>>>
>>>
>>>
>>> On Mar 29, 2023, at 11:04 AM, Jerry Shafer via RE-wrenches <
>>> re-wrenches at lists.re-wrenches.org> wrote:
>>>
>>>
>>>
>>> Wrenches
>>>
>>> Sounds like a SIP's type product, there are quite a few multi hole
>>> attachment bases that are designed for SIP's. Most will have a larger base,
>>> lots of holes for screws to attach that do not require rafters underneath
>>> to attach.
>>>
>>> Fun times
>>>
>>>
>>>
>>> On Wed, Mar 29, 2023, 8:30 AM frenergy via RE-wrenches <
>>> re-wrenches at lists.re-wrenches.org> wrote:
>>>
>>> Matt,
>>>
>>> I guess my questions are 1/2" plywood or 5/8" and how thick
>>> is the foam? Oh and I assume its nominal 2X6 T&G? Do you know how the
>>> plywood is attached?.....through the foam into the T&G?... to stringers,
>>> nailers or whatever they're called?
>>>
>>> Bill
>>>
>>> Feather River Solar Electric
>>>
>>> Bill Battagin, Owner
>>>
>>> 4291 Nelson St.(shipping)
>>>
>>> 5575 Genesee Rd. (USPS, UPS)
>>>
>>> Taylorsville, CA 95983
>>>
>>> 530.284.1925 Office/ 530.258.1641 Cell
>>>
>>> CA Lic 874049
>>>
>>> Solar powered since 1982
>>>
>>> On 3/29/2023 7:53 AM, Dave Tedeyan via RE-wrenches wrote:
>>>
>>> Hi Matt,
>>>
>>> I've used these in the past when on a comp roof and there was no good
>>> way to get into the rafters (or TJI's in this case)
>>>
>>> https://sunmodo.com/nanomount/
>>>
>>>
>>>
>>> You might want to replace the screws with something shorter and beefier
>>> though to get more grip if you are only going into 1/2" plywood.
>>>
>>>
>>>
>>> Cheers,
>>>
>>> Dave
>>>
>>>
>>>
>>> On Wed, Mar 29, 2023 at 10:38 AM Matt Sherald via RE-wrenches <
>>> re-wrenches at lists.re-wrenches.org> wrote:
>>>
>>> Hi All,
>>>
>>>
>>>
>>> I've run into a roof-mounted job where the roof is built-up with the
>>> following:
>>>
>>>
>>>
>>> Comp shingle
>>>
>>> Plywood
>>>
>>> Foam board
>>>
>>> T&G (roof deck, but also interior ceiling)
>>>
>>>
>>>
>>> The beams that hold the t&g up are wide-spaced and not convenient for
>>> fastening the full extent of the array.
>>>
>>>
>>>
>>> This being the case, I was considering other fastening options and am
>>> writing to pick the collective brain of the Wrenches to see how others have
>>> addressed similar situations.
>>>
>>>
>>>
>>> One thought I had was to use the PLP Easy Mounting Foot and I'd be glad
>>> for any opinions on that or another solution.
>>>
>>>
>>>
>>> -Matt
>>>
>>>
>>> --
>>>
>>> Matt Sherald
>>>
>>> PIMBY Energy, LLC
>>>
>>> 304-704-5943
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> List sponsored by Redwood Alliance
>>>
>>>
>>>
>> _______________________________________________
>>
>>
>> --
>> Christopher Warfel
>> ENTECH Engineering, Inc.
>> PO Box 871, Block Island, RI 02807
>> 401-466-8978
>> <http://entech-engineering.com>
>>
>>
> _______________________________________________
> List sponsored by Redwood Alliance
>
> Pay optional member dues here: http://re-wrenches.org
>
> List Address: RE-wrenches at lists.re-wrenches.org
>
> Change listserver email address & settings:http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>
> There are two list archives for searching. When one doesn't work, try the other:https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>
> List rules & etiquette:http://www.re-wrenches.org/etiquette.htm
>
> Check out or update participant bios:http://www.members.re-wrenches.org
>
> _______________________________________________
> List sponsored by Redwood Alliance
>
> Pay optional member dues here: http://re-wrenches.org
>
> List Address: RE-wrenches at lists.re-wrenches.org
>
> Change listserver email address & settings:
> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>
> There are two list archives for searching. When one doesn't work, try the
> other:
> https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
> http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>
> List rules & etiquette:
> http://www.re-wrenches.org/etiquette.htm
>
> Check out or update participant bios:
> http://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/20230331/fb692140/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: EEIlogolong.jpg
Type: image/jpeg
Size: 17117 bytes
Desc: not available
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20230331/fb692140/attachment.jpg>
More information about the RE-wrenches
mailing list