Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Combination Press Plough Plane

Testing Mortise and tenon joints

Posts

Testing Mortise and tenon joints

#1

Testing Mortise and tenon joints

Barry Irby

>I was Interested in the mortise and tenon joints that Andy was cutting in the legs of his tables. He has been doing a great job of keeping us posted on his progress. I thought the tenons would work better if they were shorter and deeper, but I did not actually know, so I thought I would test them.

Being cheap, I took and old white pine 2 x 4 that was in the shed and made two legs and two sets of rails. I cut a mortise in them at 90 degrees to each other. On one leg the joints were 3/8 by 1/2 deep by 3 1/2" long. The were the full width of the rails. The mortise intersected at the inside corner.

On the other set, the mortises were 3/8 by and inch deep and 2 1/4 inches wide. Intersecting inside the leg, causing the tenons to have a miter on the ends and leaving 1/2 at the bottom 3/4" shoulder at the top.

I cut the joints with a hollow chisel mortiser and table saw. They were snug, but rough. I slather both with epoxy and let it set for several days.

I clamped the rail in a vice and hung a sling off the leg. I added five pound lead weights until I had 35 pounds in the first one. (long shallow mortises) Before I could pick up the camera the joint failed suddenly and completely.

I repeated the setup with the deeper tenons. It held thirty five pounds and I let it sit there for a few minutes. Then I added a set of drill bits (five more pounds) It held for less than a minute and then crack/boom.

I am disappointed. I thought the glue would hold better. I think this is primarily a glue failure. Some of the wood did break, but there Si a lot of glue surface that failed. Now I am tempted to do it again with better glue. The deeper tennons did hold better and longer, but the difference is not as dramatic as I thought. I think that if you made an excellent joint with better glue that the leg would break/not at the joint on the deeper tenons. I think the top of the leg would fail on the shallow tenons,I have some pictures, but I am not good at posting them. If somebody wants to take them and clean them up and post them I will email them.


img

Re: Testing Mortise and tenon joints

#2

Deep mortice 

Barry Irby

>


img

Re: Testing Mortise and tenon joints

#3

Re: Testing Mortise and tenon joints

Kelby

>This is an interesting test, Barry. Thanks for posting this.

A mortise and tenon joint is very strong in many directions. It is a great joint for resisting forces that try to twist the leg and rail apart. But the one direction in which it is weak is the one direction I think you tested -- namely, a force pulling the tenon straight out of the mortise. When the force is in that direction, the only things holding the joint together are the friction of the fit and the glue. Those are not the things you want to count on for a strong joint.

Fortunately, I don't think small tables are likely to suffer from a force that tries to pull the tenon straight out of the mortise, so it probably isn't much of a concern in the real world. The types of forces that a small table must withstand -- such as someone accidentally kicking the bottom of the leg, or a kid climbing on the table -- create the types of twisting stresses that a mortise and tenon joint would withstand much better than I think your test results would show.

With that said, if someone were concerned about forces in the direction you tested, the solution would NOT be to simply make the tenon a little longer (as I think your test confirmed). Instead, the solution would be to employ a joint more suited for that type of stress on the joint. Better choices would be a pinned mortise and tenon, or a sliding dovetail. In your test, I think either of those joints would easily withstand very large stresses in the direction you tested.

For kicks, could I persuade you to try something? If you are willing, put that mortise and tenon joint back together with no glue or epoxy, and pin it with a 1/8" dowel (which is a pretty small dowel). See how much weight that joint can withstand using your same test with no glue or epoxy at all. I suspect the leg may break before the joint will, depending on how far down the leg you hang the weights. If the dowel breaks first, then maybe try it again with a 1/4" dowel.

Ultimately, I think it boils down to knowing which joints are better for which types of stresses. Using the right joint is much more important than using the right glue.

Thanks again for posting your test and results.

Re: Testing Mortise and tenon joints

#4

data already available

Bill Tindall, E.TN

>I think this data was posted in the Articles section. Anyway, from an academic study done in 1973

tenon length strength in in-lbs

0.5" 1700

0.75 2800

1.00 3500

2.00 5900

As you can see, strength goes up dramatically with tenon length.

Re: Testing Mortise and tenon joints

#5

Re: data already available

Barry Irby

>All my best Ideas were stolen by the ancients.

Re: Testing Mortise and tenon joints

#6

Besides what was already posted,

Keith Newton------WebKitFormBoundaryV6MZb+tiOpocz6

>I would like to see how you used your epoxy. What brand did you use, and what thickeners did you use?

If you just used straight epoxy, I am not surprised.

Re: Testing Mortise and tenon joints

#7

Ellis Walentine

Another view

Ellis Walentine

>I read Barry's test a bit differently. If the rail was clamped in a vise and the leg was cantilevered out, and the weight was hung from the leg at some distance from the joint, that would be akin to kicking the leg or falling against the table, except that Barry applied a constant load (vs. a momentary load or shock -- you physicists tell me if there should be a difference there). In a sense, such a force amounts to pulling the tenon out, but it is at the short end of a lever, so the travel of the leg is rotational about the fulcrum of the rail's lower shoulder, instead of a linear force pulling the tenon directly out of the mortise. In a case like that, as in most other mortise & tenon joinery, the staying power of the joint has much to do with the amount of glue surface in the joint. A longer tenon has significantly more glue surface, so any shearing action has to break that much more glue (or wood fibers) before the joint fails. (For purposes of this discussion, it really doesn't matter which fails first -- the glue or the wood.) But, there are also mechanical factors involved. If the tenon is fully housed, the levering force has to push out the endgrain above the tenon before the joint will fail catastrophically. That starts adding significant strength to the joint in hardwoods, less so in pine, where the endgrain has very little resistance to shearing out.

So far, this discussion has been mostly about the glue and the proportions of the mortise and tenon. The addition of other mechanical enhancements, such as pegs, will add to the overall strength, again, particularly if the wood itself has some strength and integrity, which Andy's hardwood legs and rails do.

Which brings us to the main issue: Is it strong enough as designed? You can always second-guess and/or overbuild something, but there is usually a realistic design functionlity. We all make trade-offs based on reasonable expectations. You could make a table that would stand up to just about anything if your budget is unlimited. But, realistically, it just has to sit there and look pretty, and the odds of someone kicking it or falling on it are unlikely at best.

I made nesting tea tables for a magazine article about 15 years ago, and the joints were even tinier than Andy's Shaker table joints, plus the rails were quite shallow. I used full depth tenons on those tables, and, lo these many years later, after a lot of use and abuse, the apron joints are all quite sound. The top, secured by pocket screws probably added a bit of stability to the overall construction.

So, first, I would like to commend Barry for undertaking that experiment in the interest of finding some anecdotal evidence regarding tenons and racking strength.

And, second, I think we need to understand that all joinery is relative to the design intent. We need to concentrate on what the joinery needs to do and how to properly size and proportion the components for maximum efficiency.

Ellis

Ellis

Re: Testing Mortise and tenon joints

#8

Re: Besides what was already posted,

Barry Irby

>It was Loctite Professional Extra time (60 minute) Epoxy. Bought it at the borg. Similar to the type that comes in the "two syringes" package. This came in two squeeze bottles. I added no fillers, and I only eyeballed the proportions of the mixes. Mixed several small batches and slathered every surface and assured there was plenty of squeeze out.

I don't know much about epoxy and only used it because it was what I had on hand.

The same is true for the pine I used for the legs and rails. What I had on hand that I was willing to trash for the project.

I would think you would have to make dozens of joints of several types of wood and glue them with several types of glue to test which joint is stronger as opposed to which wood is stronger or which glue is stronger.

Re: Testing Mortise and tenon joints

#9

Re: Another view

Barry Irby

>I think you "got it" Ellis. I put the apron which was 3 1/2" wide in my vice vertical, holding the leg out horizontally. The legs were about 1 3/8" square and the weight was 30" out form the joint. I picked similar pieces for the legs with approximately similar grain (rift sawn). The legs were not tapered. Could have tapered them, but my goal was to stress the joints, not break the legs. The weight had the effect of raking the joint, closing it from ninety degrees to less. Sort of like you fell against a table and lifted the near side and had the far side legs firmly caught in a carpet.

From my Wood Tech days in college I remember that many the types of wood used for framing will stand known forces of long duration, but will stand double that if the force is instantaneous. I don't know what the various forces do to the strength of glue.

I will post a picture of the load hanging from the leg. You can see the rail to the left. Unfortunately, in my hurry to snap the picture before it failed, I cropped the leg out. It was parallel to the top of the picture. The joint was not centered in the leg and the leg was beginning to rotate in the horizontal plane. IE moving toward the camera.

The way I see the forces applied to the legs in the real world is that for most of it's life the table will just sit there with almost no forces on the joint. Occasionally someone will lean on the table or push themselves up form a chair or move the table by sliding it and there will be racking forces on the joints. Sometimes perpendicular to the ends of the rails, as in my test and sometimes at various angles. Some tables will get a lot more use than others.

There will also be forces applied to the glue by changes in humidity and the pieces will expand and contract. Given long enough the joints may fail because the glue ages.

Re: Testing Mortise and tenon joints

#10

Ooops! 

Barry Irby

>


img