Wood magazine joint testing

[drumpriest]

I read an article in Wood about the relative seperation and sheer strengths of various wood working joints in oak. Very interesting as always, they came to a couple of new to me conclusions.

The winner was the mortise and tenon, no real surprise. The pocket hole, which is oft contested, did poorly. The biscuit was the worst. Double dowel did well, but not as well as M&T.

Now, the real interesting findings...

Wider mortise and tenons are stronger than than a thinner one that is deeper. So a 1 3/4" tenon that was 1" deep was stronger than a 1 1/2" tenon that was 1 1/2" deep.

Also pegging the tenon actually reduced sheer strength, causing the failure to happen at the pegs. This is somewhat intuitive, as the pegs should help with seperation, not sheer.

The dowels failed just as often from breaking the dowel than from the glue joint failing.

Thought it was an interesting disscusion. I think that the wider is better aspect of M&T might help explain the small strength loss when using floating tenons, as an integral tenon is really just of length of the board on one side, so to speak.

[cgallery]

Finally! I can't tell you how many times I've seen postings from yahoos (not here, elsewhere) that insist biscuits vastly outperform dowels due to surface area at the joint. My own experience had been quite the contrary.

[LCHIEN]

Keith, is that article on-line? I've read it before but would like to see it again. Can you post the URL?

CG, I think the main reason people use biscuits over dowels is the convenience of making the joint, not necessarily the strength. In terms of strength, I'd rate them both as "adequate".

[cgallery]

Loring,

I agree that the vast majority of users prefer biscuits due to ease of use. There are a certain # of users that drink the manufacturer Kool-Aid and keep insisting biscuits are stronger. Not just a little bit, but "way way stronger." Here is an example:

Elmer, where did you read this information about the strength? I have a book that compares the strength of biscuits to dowels and the load bearing strength is far superior to that of a dowel in a joint. The biscuit distributes the downward pressure over a 2" area. The dowel distributes it over a 3/8" area if using 3/8" dowels. The dowel may not break but the board it is glued into will. The dowel forces all the pressure on the board into one spot. The board you are gluing will fail with the dowel long before the biscuit. You have 2" of 1/8 inch material vs. 3/8". These are the printed facts. Taken from "Biscuit Handbook" by Hugh Foster. I was skeptical also until I bought the book and read it. 10 year later, no failures with biscuits.

That said, there are definitely applications that dowels are better but not as many as there were.

So I guess you can't believe everything you read. Direct testing is the way to go.

Finally, while I find dowels can be tricky to use (at least initially), I actually prefer them over M&T for one reason: Once you master dowels, you get a very good, consistent joint. I've struggled w/ M&T joints. I agree that tight fitting M&T joints are stronger than dowel joints. But I think it is FAR more difficult to get ten tight M&T joints than 10 strong dowel joints.

[Knottscott]

I vaguely remember reading a similar article some years ago that pretty much came to the same conclusion....can't remember which mag, but it was prolly 5 years ago.

I find I'm using my biscuit cutter less and less. It just doesn't add enough strength to bother with, b/c most of the glue joints are stronger than the wood already. If it needs reinforcing I go to pocket screws first, or will do an M&T joint if really needs to be strong.

[cabinetman]

I was using splines and dowels before biscuits came out. Nothing wrong with biscuits for certain applications. A spline joint can be considered as a loose tenon. With the variety of choices of material, it works well. I still have faith in splines and dowels, but which ever one I choose I don't use them for alignment. The fit and positioning should be achievable when clamped up with glue. Dry fitting is a failsafe. Done right a MT is a viable joint, for applications such as chair construction.



"I'M NEVER WRONG - BUT I'M NOT ALWAYS RIGHT"

[Ken Weaver]

Seems to me that one question dominates - "how strong do I need the joint to be?" On a table top for example sheer strength doesn't seem to be an issue and therefore using biscuits as an alignment aid makes sense. I would guess that there's room for all of them.

[Jim Boyd]

It was my understanding the biscuits were for alignment, not strength

[Crockett]

You're right Ken,the article mentions that point. If you are making something that won't typically come under the stress that the machines they used in the test create, you will probably be fine, even with butt joints in some cases. It is a great article that will help people make decisions when building items that may be "stressed" for one reason or another.
Al

[LCHIEN]

Oh, I saw it - it's in the current issue of Wood Mag that just arrived.

[cabinetman]

Jim

After you use them you'll find out you can't depend on them for alignment. The cutter allows just enough for glue, that also allows for uneven edges if you don't use cauls, and proper clamping procedures.



"I'M NEVER WRONG - BUT I'M NOT ALWAYS RIGHT"

[JimD]

I should have my copy of the most recent Wood by the time I return from the trip I am on at the moment. Will be interested to see what they have to say. FWW tested years ago and came to different conclusions (mortise and tenon best but biscuits were second with dowels somewhere later (I don't remember pocket screws in the test) - they tested door corners). My own direct experience is also better with biscuits than dowels for doors. Still, I will be interested in the results.

Jim

[Tom Slick]

really what the test shows is that you need to use the correct joint for the job, there is no one size fits all.

[25]

I just picked that issue up last night. I haven't had a chance to read it yet, just looked at the graphs so far. I was rather surprised how much the pinned M&T joint weakend. Was it glued or only pinned? If it was only pinned then of course the dowels are going to be the weak point.

[JimD]

25 - the mortise and tenons were all glued. If you look closely, you will note that several of the mortise and tenon joints were 3 inches wide while all the other joints were 2 inches wide. I think Wood was right about the pegs being decorative.

For those that like dowels, I agree that they came out well in these tests. I think that Woods test indicate that dowels are good joints for a face frame but don't tell us a lot about joints in pieces of wood wider than 2 inches - at least with respect to biscuits. A #0 biscuit is the smallest commonly available and that is what Wood tested because they were working in 2 inch wide pieces like you would use for something like a face frame. In addition, the shear test that Wood did pulled the grain apart in the little biscuit but worked across the grain for the dowels and the mortise and tenon joints. Not unrealistic but it would be interesting to see how the results would change if they tested in shear along the flat part of the wood instead of the edge. My biggest disappointment is that Wood tested 3 inch wide mortise and tenons but didn't test #20 biscuits that would work well in 3 inch wide stock and have about twice the glueing area as compared to a #0. It would also be significant to see what effect using double biscuits would have. If you test 3 inch wide mortise and tenon joints why not more than one option in biscuits? It would take a lot less time to test double #20 biscuits in 3 inch wide stock than it took to make those mortise and tenon joints.

Overall I think the Wood article gives us a thumbs up for dowels but I'd like to know their diameter and how the holes were drilled and if they were plain or grooved or what. I do not think it was a very fair test for biscuits except it tells us that if we are using wood that is narrow and can only take a #0 biscuit we should not expect much strength. Dowels would work better (as would pocket screws). I think I already knew that.

Jim