Sawdust on the Floor

A recent article in Fine Woodworking magazine caught my eye for several reasons. The first was the subject matter, the strength of woodworking joints, and the second was the method of testing.

Fine Woodworking commissioned an independent lab to conduct testing on most of the common woodworking joints used in frame joinery. To control the test results, all of the joints were made from Cherry, and glued with Titebound III waterproof adhesive. Each joint was clamped for at least one hour, and allowed to cure for five days before the joints were tested.

The joints were then subjected to a controlled compressive load in a Servo Hydraulic Test system. This is basically a hydraulic actuator that presses down on the joint at a controlled rate. This method of loading the joint, was designed to simulate what is known as a racking load. Racking force or racking load is the force that causes what started out as a rectangle to become a parallelogram. Or in other words it is the load felt on the non supported side of every door. The joint is what keeps the frame squared up, but as soon as the joint fails the door will droop or fall away from the supported side. This same racking force is also seen on top of table legs where they join the frame of a table, when the bottom of the leg is struck. When the racking force is great enough the joint fails. As each joint was tested the actual force or load felt on the joint was recorded by the servo hydraulic system.

So what about this article peaked my interest? Well, for one, the results of the test were surprising. Some joints scored much lower then I would have suspected, others were much higher. The other reason is that I happen to work for the manufacturer of the Servo Hydraulic Test System used in this test. It was really neat to see something you have worked on end up in print.

The results of the test are listed below, what surprised me was that the plain and simple half joint had the highest peak load, while the much touted cope and stick joint had a low peak load. Does this mean that we should all abandon joints like the cope and stick, or the Festool Domino for the half lap? Of course not, the half lap and bridle joints are very strong but let’s face it, they are not the most handsome joints.

I do think that having this information will help each of us make more informed design decisions when it comes to selecting frame joints. If you are building a door that will end up with a very heavy insert , like a piece of plate glass then I would tend to shy away from the biscuit or the Domino joints in favor of the 3/8 ” Mortise and Tenon or the splined miter. But if the door were to have a simple raised panel insert, then the traditional Cope and Stick joints make sense.

We all have to balance the look and feel of the project with the mechanical properties of the joints we are creating. Building a piece that is going to just be displayed versus table with a heavy top, or one that will get daily use require different types of joints.

Joint Peak Load At Failure

Half Lap 1603
Bridle 1560
Splined Miter 1,498
3/8″ Mortise and Tenon 1444
3/8″ floating Mortise and Tenon 1396
Miter 1374
3/8 ” wedged Mortise and Tenon 1210
3/8″ pinned Mortise and Tenon 1162
5/16″ Mortise and Tenon 988
Beadlock 836
Dowelmax 759
1/4″ Mortise and Tenon 717
Pocket Screw 698
Domino 597
Biscuit 545
Butt 473
Cope and Stick 313
Stub Tenon 200

Related posts:

1. Woodworking Techniques – Mortise and Tenon Joints
2. Woodworking Tools Festool Domino Joiner