We’ve been installing metal roofing attaching the panels with screws to 5/8″ plywood roof sheathing. Infrequently, some of the screws will back out over a period of time (4-5 years). Usually the exposure of the roof is to the south but not always and the pitch of the roof varies from 7/12 to very steep ie.26/12. Does anyone have a guess as to what is causing this? and any remedies that you have come up with would be greatly appreciated. Thanks.
Dennis Cozzetti
Replies
Just a guess based on similar experience.
This is going to sound weird, and will probably draw some flak, but it happened.
We get a real lot of freeze/thaw cycles here. Like daily for about 4 months every year.
A friend of mine is a metal roofer and had a big contract, about a dozen buildings. The first one, by Spring all the screws are just a little bit backed out. Close observation on his part, and he determines that the place is unvented and the moisture is freezing inside the roof on the screws and basically ratcheting them out a little every day.
He goes to the project manager with this theory, the PM decides to go along with it, vents the roofs on the rest of the buildings, and there have been no problems with any of them.
Weird? Sure. I warned you. <G>
DRC
Dave, that's way to complicated. It's Elves. They come at night and break s**t. Happens all the time. Who do you think makes dead batteries & stuff like that? It's Elves.
Same guys that unscrew light bulbs and hide your keys at night? Elves, Dave, Elves... Joe H
I have two in my house. One is called Idunno and the other is called Notme.
Week, I had three. All grown & gone now, s**t still breaks. Joe H
Thanks for the advice, bu the roofs in question are all vented with freize boards along the eaves and a continous ridge vent at the peak.
OK, elves it is.
It explains a lot, actually.
One of these days I'm going to wake up a few minutes early and catch the little b@st*rd that is responsible for all this stuff.
DRC
Dennis,
I see this sometimes also, think it comes from the metal walking. I remove the screw, squirt in some Lexel, (not silicone) and put a new screw in. Might have to upsize the screw.
KK
Edited 8/15/2002 4:19:25 PM ET by kkearney
I suspect that the cross section of the roofing is relaxing away from the screw heads. The screws are held tight to the raised portion of the roofing in opposition to the tension of the metal deformed to make the raised profile. It seems possible to me that heat cycling would make this happen more quickly.
4lorn, are you suggesting that hese might be installed on the ridges instead of the field where they belong?
I can't answer the question with any certainty but I'll throw out a little theory for you all to volley back and forth across the net...
I'm suspecting some kind of action where the metal surface is moving and causing a screw or two to unscrew over multiple cycles. Thermal expansion/contraction is the likely suspect.
My method for installing metal roofs was to pile up the required number of sheets for a roof side on pallets and do my layout on the ground. Used a scratchall to divot each spot and them pre-drilled all holes at once. sometimes twenty or so sheets. Then all the screws line up nice and straight across the roof. Other than neat appearance, there were several other advantages to just jab-starting them on the roof. I used a drill bit that was slightly larger than the threads on the screw shank. That way, the metal is not bound too tight to allow for thermal expansion. It is also easier and safer to start the screw while balancing on the roof.
I had to replace several roof sheets that were installed by a competitor. The snow and ice. It may be that the screws were overdriven or that cheap screws were used. Some signs of being overdriven were present, such as neoprene washers compressed to the point of splitting. Not too many available for observation tho'
I believe that at least part of the reason for this failure was that the roofs were installed in the heat of summer and that the screw holes were not predrilled to allow for any movement.
Part of the reason for the ribbed design of the roof panels is to allow those high spots to take the movement laterally but some must still transmit lengthwise. I can imagine this movement lossening a few screws.
Personally, the only ones I have seen do this were those that happened to hit a seam or a void in the plywood, or those that were overdriven.
Hope this helps.Excellence is its own reward!
Not being a roofer this is not my area of expertise but from observation of roofs installed on projects I have worked on and roofs around town I conclude that around here 5V and corrugated roofing is screwed through the ridges except at the drip edge where two extra screws per flat are added to prevent wind uplift. They are screwed on the ridge to keep water from leaking around the screw holes by keeping the water flowing away from the screws. I didn't think that there was any other way.
Raised ridge roofing is installed with strips screwed to the deck and folded into the joint. That way there are no screw holes to leak.
All manufacturers of the ribbed metal roofing instruct tha tscrews should be placed in the flats/field portion of the profile. It used to be that what you explain was the correct method for installing old corrugated sheets with leadheaded ringshanked nails. The screws are a whole 'nother animal. If you put the screws in the tops of the ribs, it is impossible to draw them tight enough to seal the neoprene gasket. That is, unless you mash the rib down so far that you are creating a buckled uplift in the field. Now you have added wind uplift to wiggle the panel and loosen whatever little bit of seal existed at the screw. Add to that the fact that the shank of the screw is being made to flex. Screws being more brittle than nails, it won't be long before they break off.
The design intent of the modern tech screw is that the metal cup washer under the head will snug down over the neoprene washer. The curvature of the concave washer presses the softer neoprene in towards the shank (where the hole is) and down toward the metal (where the water runs) at the same time. You need the wood surface (whether strappingm purlins, or sheathing) behind the metal to draw against for this to happen. screw guns with adjustable clutches let you set these with just the right amt of tension easily. Using a regular drill/driver with no clutch malkes it hard to get it just right and many will be overdriven to the point that the neoprene is mashed out from under and split wide open, creating a channel for water to reach the hole. The washer protects the neoprene from the UV rays of the sun so properly driven screws will last a good twenty years.
All of this is clearly explained and diagrammed in the instructions available with roofing products. Instructions are good things to read before applying/installing stuff in general.
Send your donations to;
Mythbusters Galore!
:>)Excellence is its own reward!
Hmm. Interesting. Might be a few roofs around here installed incorrectly. I didn't install any of them, being an electrician, so I don't feel bad. From what I heard none of them leak. Maybe in time. Your comment about manufacturers directions is, of course, the way to go. Within a year or so I might be installing a roof on a house I own. I always liked galvanized, which is why I notice the details, and will be careful to follow the manufacturers directions.
Just goes to show you that things change and popular wisdom in any particular local can be wrong.
I knew I wasn't picking on you, old buddy.
Another common mistake guys make by not reading the instructions is this,
They put the female side of the lap on top instead of under. In this sketch attached I try to show it with the male as a blue line and the female in red. You can see that the toe edge of the red sits on the sheathing for support and there is a little void formed by an extra break in the rib. The purpose of it is so that any water that wieeps up under the lap by capillary action or wind force has a place to run without backing up over the lap and leaking into the roof.
Some folks reason that if laid backwards from what is show, they get more coverage at the lap. Don't work that way though.Excellence is its own reward!
oops here's the attachmentExcellence is its own reward!
I'll try to keep that in mind. Good drawing.
I agree with you that thermal expansion is probably the most likely candidate. We install the panels the same way as you do so that the layout is nice and straight. On the next roof, I'll try a larger diameter bit for the pilot holes. We also adjust the clutch on the cordless drills to make sure we don't overtighten the screws and follow the manufacturer's instructions as well. As far as I can see, we're doing everything like we should. Maybe it is elves, bored elves.
Dennis
all else done right, I suspect a football void in the plywood to contribute.Excellence is its own reward!
Are you expecting screws to hold in 5/8 ply?
I'm sorry, I haven't had the good fortune to do a metal roof, but isn't that expecting a bit much?
Screws in 5/8" CDX is typically successfull. Why would you suspect otherwise, Qtrmeg?
Excellence is its own reward!
I dunno, that is why I asked.
This screwy concept in my mind says metal roof panels move a hell of a lot more than claps, and you have to hit a stud to keep a clap on a wall.
Just curious is all, just something else I know nothing about.
Screws Backing Out Solution
I have seen this problem. This is the fix.
WebSite for hardware: http://www.mutualscrew.com
Order A box of: SKU: 0816ABWSKC
8 X 1 Indented Hex Washer Head Steel Zinc Plated Unslotted Serrated Type AB Sheet Metal Screws
Order a Box of: SKU: FSSW08
#8 X 1/2 Outside Diameter Stainless Steel 18/8 Bonded Neo EPDM Sealing Washers
Take old screws out, throw away.
Replace with 1 new washer &1 new Screw from the screws you ordered from Mutual screw or a suitable match
THE SERRATED HEAD IS A MUST!!!!!!!!!!
Do not over tighten!! Meaning, don't mash the rubber washer out from under Stainless Steel washer.
Set the torque on your drill, if your a brut, try a little finesse or use a hand driver
If the old hole is to sloppy for a #8, order #10 instead.
This will solve your problem. I live in New Mexico, the elements here are, wind, extream heat, extream cold. I experienced the same problem. This is working for me. I install new metal roofs using this Technique and it is a much better way, as I have not experenced any screw backing out problems.
All the best.......... Lou
Do you realize you're replying to a question from 2002?
large cycle changes in humidity have the same effect as freeze/thaw cycles, even more so in softer woods like hemlock.