Rules for Specifying Woodscrews

[John Baum]

Has anyone seen a reasoned discussion outlining rules of thumb for specifying wood screws. Clearly there are screws like a #12 that is 1/2" long that approach being ridiculous. Likewise a #2 screw that is 1/2" long has a likelihood of breaking under load.

Today I am faced with specifying a 5/8" long wood screw to draw together a lapped joint in birch. A #6 screw gives me 18 threads/inch, a #8 gives 15, and a #10 gives 13. Assuming the tip of the screw penetrates 5/16", I have 5.6, 4.7, or 4.1 threads engaged. In the process I trade off a smaller root for more threads or fewer threads for a larger root diameter. I'm interested in an engineering based rule of thumb for specifying the screw size for a given length.

Thanks

baumgrenze

[pat warner]

Way too many variable to generalize. Material, its EMC, the tensile requirement, tpi tolerable, your drilling skills, drill choices etc. origin of fastener, head configs, fully threaded or partially, hardened, sheet metal, wood, brass, ss, carbon steel; you get the pix.
Make your best judgement and test on scrap. I study drilling & fasteners (in wood, plastic and aluminum) daily.
Charts are rough guide lines only. A 5' emprical test with identical fasteners, drills, stock and a big driver will give you real and practical values.
A good question.

[Erik Christensen]

Man reading some of the posts on here makes me feel like an ignorant rube - that level of detail would never have occurred to me. I just hold up a screw to the 2 pieces to be joined to make sure it won't come through the show side and go "yah that'l work". Now I am getting flashbacks of my time in a navy nuclear power plant where we would get ding'd post work for "insufficient thread engagement"..... yikes just another thing to worry about when i am in the shop and feeling like I am not doing something right :)

[Tom Walz]

For a question such as this I would consider posting in Engineering Tips as well. http://www.eng-tips.com (http://www.eng-tips.com/index.cfm) Maybe one of these two forums: Welding, Bonding & Fastener engineering (http://www.sawmillcreek.org/threadminder.cfm?pid=725) or Wood design and engineering (http://www.sawmillcreek.org/threadminder.cfm?pid=337).

[Steve Milito]

I'm with you. I go with the Goldilocks approach, what do I have, does it look to big, does it look too small, is the right drill bit at hand.

[Chris Friesen]

I recommend chapter 7 ("Fastenings") of "Wood Handbook -- Wood as an Engineering Material". It's available online at http://www.woodweb.com/Resources/wood_eng_handbook/Ch07.pdf

Interestingly, in some hardwoods (maple for instance) the strongest joint will actually be to use machine screws going into tapped holes in the wood. This gets you accurately cut threads in the wood, giving better holding strength and it also gives a wider set of options for the thread pitch. It also means full holding power the full length of the screw rather than losing strength as it tapers to a point.

For softer woods I'd tend to go with a low-root thread with large sharp threads. I happen to like Robertson screws, though the Spax ones are also very good (I'm sure there are others, that's just what I've actually used.)

[Steve Milito]

I recommend chapter 7 ("Fastenings") of "Wood Handbook -- Wood as an Engineering Material". It's available online at http://www.woodweb.com/Resources/wood_eng_handbook/Ch07.pdf

Interestingly, in some hardwoods (maple for instance) the strongest joint will actually be to use machine screws going into tapped holes in the wood. This gets you accurately cut threads in the wood, giving better holding strength and it also gives a wider set of options for the thread pitch. It also means full holding power the full length of the screw rather than losing strength as it tapers to a point.

For softer woods I'd tend to go with a low-root thread with large sharp threads. I happen to like Robertson screws, though the Spax ones are also very good (I'm sure there are others, that's just what I've actually used.)
I particularly like page 7-8.

measure twice
calculate the hyperbolic sin
square it

:)

[Chris Friesen]

I particularly like page 7-8.

measure twice
calculate the hyperbolic sin
square it

:)

It does get a bit involved, doesn't it? I've got an Engineering degree and I still just sort of skim over most of the formulas and just look for easy-to-use tables. :)