18 May 2010

Cabinet making - 5

This is part of a series of posts, inspired by the New Yankee Workshop (NYW) presentation Oak Bathroom Vanity - Program #106, about alternative cabinet design and tools use.

After replacing mortise & tenon with biscuits previous post, I look how to improve the strength of a biscuited frame.  Biscuits are interesting but are often discredited as being to weak.  The fine woodworking magazine (FWW) did a joinery shoot-out in January 2009. It gave values in the following ranges :
  • half lap: 1600
  • splined and standard miter: 1500 - 1400
  • mortise & tenon: 1450 - 700 (depending of the thickness)
  • stub: 500  
  • dowels - pocket screw - biscuit - stub tenon: 700 - 200 (shallow joints where the style splits along the grain)

Shallow joints
For a dowel joint it is easy to increase the depth of the dowels. I buy dowels by the meter, so no problem there.  The people of  dowelmax showed (in reaction to the FWW article) that 2 inch deep dowels avoid that the style splits along the grain and allow for strong joints. I am not sure if all the dowels need to be lengthened as the exterior dowels are the most critical .

I implemented the same idea but for biscuits this time.  I added a half dowel (a dowel drilled cross-grain, flush with the surface, only there to avoid splitting) close to the end of the stiles. The next step is to make a standard biscuit joint, cutting through the top of the half dowel. With this I made a number of test pieces with a normal biscuited joint (nr 0) on one side and a biscuited joint combined with a half dowel at the other.  When hitting the test piece with a mallet, the normal biscuited joint failed easily. When I tried to break the reinforced joint I had to take a bigger hammer to make it break.

The pictures show
- the tested joints with the wood splitting at the depth of the biscuit of the reference joint, where the strengthened joint does not break
- a cross-cut of the biscuit and the dowel
- same with also a half dowel before cutting the biscuit groove

For pocket holes the half dowel can be replaced by a narrow headed screw,  making it a screw (and glue) only solution.  But I did not test it yet.
For stub tenons half dowels are probably a drastic improvement.

Miter joints
Miter joints show in the test a very decent score.  Although my guess is that they are asymmetrical in resistance, being at their best when closing the joint, what was tested by FWW. And shallow joints are probably at there best when opening the joint.
Where a simple miter joint showed a score similar to a splined miter joint, at the JLC forum I saw examples where in adverse condition (high moisture) there is a clear differences between both methods after a few months.
Biscuiting miter joints has an added advantage as the diagonal offers more room for large size biscuits.

Using miter joints for door frames

I have an example where a mitered (drawbored mortise and tenon) joint was used for a door frame. The main reason is probably the use of profiles.  Apart from the top with its curved profiles, everything is made with continuous profiles.
This makes it possible to build the door in three distinct steps (possibly different carpenters or workshops):
1. Cut the wood and make the panels
2. Make the profiles
3. Cut the joints and assemble the door

To clarify the style: This oak bonnetière was assembled in the fifties, it is unclear to me how much older the door is or what the style of the original door was.

That's it,  next time I look at the top in a last episode.