Joel D said: "I would be concerned about the whole roof lifting up in a strong wind"
Me, too. http://www.youtube.com/watch?v=IHxc8AwG5Ac This is not a tornado... Not a hurricane... Just wind. And not a particularly strong wind, either. Helps explain why TX has built so many wind farms... I really like the S-5! in the right circumstances. I am concerned that it's so "easy" that it can be and has been used in applications where it has been vetted about as well as McPain's hunting mate. In nearly all evaluations I've done of existing S/S roofs, where the structure was actually able to support the added weight of an array, the weakest link has been the attachment of the roof skin to the building. Uplift forces at the roof surface (array location) are transferred to the building structure via 2 attachements, not just 1. The "clip" has a top and a bottom. Both ends must be evaluated. I've found more than a few that don't pass a responsible structural analysis "as-is"... Without solar. Older documentation from metal bldg/roof mfrs provided a psi or psf rating for the "top" connection... Where the clip attaches to the metal skin. "Snap-Click". Be sure to double-check the units... I've seen psi published. There are 144 sq inches in a square foot... If you're looking for 35 psf and you see 350 psi, you don't have a 10X safety factor! What was less often published is a rating for the attachment of the clip to the structure. This gets really squirrely when the clip is screwed into OSB &/or plywood. Another point that I have seen engineering calcs fail (but not physical failures so far), is tear-thru... Where the clip is torn right over the screw head that holds it to the structure. This is different than "pull-out", where the screw pulls out of the structural member. Frankly, I'm not convinced that the couple of times where the engineers came up with this it could possibly happen... I don't think the top of the clip is gonna hold on long enough to tear the bottom of the clip this way. But that's what the guys with the fancy initials and the little round rubber stamp said... Hopefully your application will have complete published data on the system used and conditions found at the site. Don't forget to actually VERIFY the clip spacing and attachments! Again, metal over OSB/ply requires special attention. In these cases, I recommend taking some of it off in a few areas to investigate if you can't otherwise prove what's in there. One of my favorites is the wood-truss/osb install where the original documentation and engineering is based on (2)ea x 1-1/2 or 2" screws directly into the top chord of the truss at each clip with 48" spacing up the truss. From the ground you can see that the seams don't line up with trusses anywhere and you can clearly see the tip of a single screw come through the OSB by a whopping 1/4" in the eaves. Sooooo... You want me to believe that you "did it the right way" where nobody can see it, but not where it's in plain view?... I've been to Missourri... Try again. Oh, wait... That's not my favorite part! My favorite part is breaking the news to the owner that his roof is gonna blow off so he can't have solar on it unless we do standoffs. And, yes, I TOTALLY appreciate that he "built it this way so he could have solar and no penetrations". (Just kidding folks. This is NOT my favorite part. These situations break my heart and tick me off to no end. Somebody cheated and now somebody else has to pay for it... One way or the other) Look out for roof-skins that are already coming loose from the structure. In my experience, these situations generally occur where there are long &/or multiple panels between a "ridge" and an "eave". I haven't personally observed it in flat-slope apps, but suspect it happens on those, too. Be VERY cautious on steel buildings with open walls. I've seen plenty where the skin is already flopping loose when you get there. The clips have already released from the skin and/or completely deformed. I've mostly seen these in ag applications. Riding arenas, hay-barns, tool sheds, etc. A municipal dump transfer station building, too... That one was probably the worst of all in this regard. Here was a beautiful, long, wide-open roof that was literally flapping in the breeze as you walk on it. The pressure differentials on these buildings are magnified greatly from those of a normal "enclosed" structure... Positive pressure below the roof + negative above = lotsa lift. Probably the single biggest factor that goes into causing these to fail is that wind is not "smooth and steady". If you are the roof on one of these buildings, you are being pounded continuously. Stand on or under one of these buildings when there is a 10 mph or greater wind and just listen to it... It doesn't just "creak and groan"... It "bangs and slaps". Another sign of trouble to watch for is horizontal creasing of S/S panels. In order for a panel to crease, it has to get bent some time... Either before or after it is installed. When these profiles get bent, the seams get distorted. They don't fit like they were designed to. Since the clip design is dependent on a perfectly shaped seam, you kinda gotta wonder if the "new shape" is gonna hang in there when mother nature starts pulling on it... That dump transfer station roof had these creases all over it... Don't know which came first, the crease or the loose roof, but at this point it doesn't really matter, does it? I'm comfortable with the holding power of an S-5! For purposes of wind analysis, it's the strongest part of the attachment system. For applications where this method is an option, right after making sure the building can physically support an array, the next thing I look at is the roof-to-structure attachments. I never kept an actual tally, but I'd estimate that only about 20% of the potential S-5! projects I looked at over the years passed this phase of the evaluation without requiring modifications of some sort. For "smaller-ish" arrays, don't waste $ on rails. For "normal" sized, rectangular, shaped modules, turn 'em sideways (landscape). Figure on using an S-5! at every seam or 24", whichever is greater, and run the math that way. I, personally, want a minimum safety factor of 2+ for these apps. In this part of the country, it is common for all-steel buildings to have only 1 clip for every 12-14 sq ft... Horizontal purlins @ 63-66" spacing with vertical seams every 24 or 32"... You MIGHT get this to calc in zone 1 of a barn with low-slope roof, 85 mph basic, cat B exposure, good screws, high-quality roof profile, and a strong clip... I've seen it done. But it doesn't get near a 2X safety factor... Keep in mind the "S-5! WARNING... The S-5! clamp is a handy gadget for a great many uses, but will not perform miracles." < http://www.s-5.com/Home/index_457.cfm > Miracles is what hammocks is made for... Matt Lafferty [EMAIL PROTECTED] _______________________________________________ List sponsored by Home Power magazine List Address: RE-wrenches@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