wood fasteners - research results
nic...@nicad3.nic.bc.ca
regarding wood fasteners. In the winter 93 issue of WoodCuts,
a magazine put out in Canada by Lee Valley Tools of Ottawa, they
presented the results of a study that they carried out with the
aid of the Materials Testing Lab at Algonquin College, also in
Ottawa. The study was done to determine the holding strenght of
machine screws vs wood screws. The results were quite interesting,
since most woodworkers wouldn't normally consider using machine
screws in wood.
Briefly, their study compared the holding strength in hardwood
(oak and maple)of brass machine screws and brass wood screws
in sizes 6, 8 and 10 (6-32,8-32 and 10-24 for the machine screws).
The wood screws all had CUT threads as opposed to rolled threads.
The tests covered both Tension(along the axis of the screw)
and Shear (across the axis of the screw). The machine screws were
inserted into threaded holes, tapped into the wood with standard
metal cutting taps. The wood screws were inserted into pilot holes
of optimum size. (The minimum size that allowed the screw to be
driven to its full depth without deforming the head.)
In all cases, where the depth of the hole was greater than 1/2",
the machine screws failed due to metal fatigue at about 666 lbs
of tension, the wood screws pulled free of the wood at much
lower loads. I have summarized their graph in the table below.
Hole Failure Tension in Lbs
Depth Mach.Screw Wood Screw
1/4" 195 35
1/2" 521 270
3/4" 666* 404
1" 666* 464
* indicates failure of the metal
all testing done with #6 wood & 6-32 mach. screws in oak inserted into the
edge grain.
There was also some testing done of optimal pilot hole sizing for screws
with the following results.
Screw Type Clearance Pilot
Hole Hole
#6 9/64" 1/8"
#8 11/64" 9/64"
#10 13/64" 5/32"
6-32 3.5mm 2.5mm
8-32 4.5mm 3.2mm
10-24 5.0mm 3.5mm
(don't ask me why they switched to metric!)
Wood screws faired better in end grain insertion tests on oak,
it didn't seem to make much difference in the maple. Failure in end
grain insertions in oak was due entirely to catasrophic failure of the
wood itself. They recommend that you use a slightly larger pilot hole
in end grain with wood screws to avoid graing separation. The general
recommendation is to use wood screws in end grain, as the tapping of
the threads for the machine screws tends to shear the wood fibers, hence
weakening the thread. Using an oversized pilot hole in end grain results in
a joint that is about 75% as strong as one in edge grain.
In most circumstances, brass wood screws are stronger than their machine
screw counterparts, but don't hold as well under load.
The minimum thread engagement in any circumstance should be 1/2" if there is
any chance of tension being applied. Also, in insertions of 1/2" or less,
machine screws will hold much better.
The following table indicates the breaking strength and recommended insertion
depth of the fasteners tested. (the point where the pull out resistance is
equal to the breaking styrength of the fastener.)
Screw Avg Breaking Recommended
Type Strength (lb.) Thread Engagement
OAK MAPLE
6-32 670 5/8" 1/2"
#6 660 1" 1"
8-32 1030 1" 1"
#8 1155 1&1/4" 1"
10-24 1310 1" 3/4"
#10 1355 1&1/4" 1&1/8"
The above is condensed from 4 full pages of information in the article (there
aren't any ads in the magazine.) So if its a little disjointed, please
forgive my paraphrasing.
This is presented purely for information, I don't claim to know which is
better, just that I will now use more machine screws for edge grain fastening.
john nicklin
PS. If you want a source of very good (maybe the best) hand tools, get a Lee
Valley catalogue.
David Thompson
>For general interest, I would like to pass on some observations
>regarding wood fasteners. In the winter 93 issue of WoodCuts,
>a magazine put out in Canada by Lee Valley Tools of Ottawa, they
>presented the results of a study that they carried out with the
>aid of the Materials Testing Lab at Algonquin College, also in
>Ottawa. The study was done to determine the holding strenght of
>machine screws vs wood screws. The results were quite interesting,
>since most woodworkers wouldn't normally consider using machine
>screws in wood.
>
From the information provided, I infer that the tests were
done on *freshly made* drilled/tapped holes. I can imagine machine
screws failing more quickly than wood screws, over a longer period
of time, due to wood movement, because of their shallower threads.
This is pure conjecture, however!
Dave Thompson \\\|||/// d...@deimos.caltech.edu
\ ^ ^ /
< O O >
-----------VVVV------- V -------VVVV------------------------
One cannot attain the limit of craftsmanship,
And there is no craftsman who acquires his total mastery.
- Ptahhotep, c. 2350 B.C.
But I can try!
- me, 1993
------------------------------------------------------------
nic...@nicad3.nic.bc.ca
> In article <1993Jun2....@nicad3.nic.bc.ca>, nic...@nicad3.nic.bc.ca writes...
>>For general interest, I would like to pass on some observations
>>regarding wood fasteners. In the winter 93 issue of WoodCuts,
>>a magazine put out in Canada by Lee Valley Tools of Ottawa, they
>>presented the results of a study that they carried out with the
>>aid of the Materials Testing Lab at Algonquin College, also in
>>Ottawa. The study was done to determine the holding strenght of
>>machine screws vs wood screws. The results were quite interesting,
>>since most woodworkers wouldn't normally consider using machine
>>screws in wood.
>>
> .. results deleted
>
> From the information provided, I infer that the tests were
> done on *freshly made* drilled/tapped holes. I can imagine machine
> screws failing more quickly than wood screws, over a longer period
> of time, due to wood movement, because of their shallower threads.
> This is pure conjecture, however!
>
at the time the wood test blocks were made by Lee Valley, which I would assume
would have been several weeks at best befor the test was done. Your conjecture
raises an interesting point, are a relatively few deep threads (on the wood
screws) better than up to twice as many slightly shallower threads (on the
machine screws). I would imagine that much of the holding power of the machine
screws comes from their shape (a cylinder) whereas the wood screw is
essentially a cone. (my conjecture).
Over to you.....
john nicklin
David Thompson
.. stuff deleted
>The drilled and tapped holes for the machine screws were apparently done
>at the time the wood test blocks were made by Lee Valley, which I would assume
>would have been several weeks at best befor the test was done. Your conjecture
>raises an interesting point, are a relatively few deep threads (on the wood
>screws) better than up to twice as many slightly shallower threads (on the
>machine screws). I would imagine that much of the holding power of the machine
>screws comes from their shape (a cylinder) whereas the wood screw is
>essentially a cone. (my conjecture).
>
>Over to you.....
>
>
>john nicklin
Again, because of wood movement I would think that in the long run
(years as opposed to weeks) the wood screws would do better in wood than
machine screws. Something else I forgot to put in my previous post on this
thread is this: how often would you design something to hold 100 lbs (let
alone 600!) with a single 6/32 screw? I admit I tend to overdesign things
(an aluminum lens + CCD mount I made is held together with 8 1/4-20 bolts
though it only has to carry about 30 pounds!). If you were to make a chair,
for example, with 2 wood screws on each major joint (yes, I agree - mortise
and tenon, mortise and tenon, but that wouldn't serve to illustrate my point!)
then you have some 800 pounds of holding power *at each joint*! Since properly
designed chairs *distribute* the forces encountered in normal use, I doubt
that you would ever exceed the ratings on the screws.
I think that the tests performed as described were *static* tests.
Perhaps a better test to do for the machine/wood screw comparison is a
dynamic one. Something like dropping weights, swinging pendulums, or exer-
cising a joint by repeated flexing may be more indicative of the relative
merits of machine vs. wood screws in wood, especially in the long run.
Cheers,