Kumiko Sampler - #3 Making Slats & Rails

With the jig all set up spending some time at the table saw ripping the lattice slats is next.  Doing all the cutting went pretty fast.  I bet between 10 and 15 minutes is all it took to get 36 ripped pieces.  I ended up working from both sides toward the center.  It just seemed like that’s where most of the flaws were and I wanted to get the maximum usable footage out of the blanks.  As you can see there is still an uncut blank about an inch wide in the stack.  That’s for another set of pieces I will get into in just a bit.

Alder does not have a lot of grain and I think that the lattice slats will be mostly interchangeable.  However, just in case I marked the ends of each group so I could keep track of which slat came from which board.  This probably won’t be needed but trying to put them back in order after getting scrambled will be a lot easier with the marks.

The basic framework for the Kumiko is a gridwork of half lapped rails.  Here is an exploded view of the ones I will need for this project.  The long ones are 9 5/8” long and the short ones are 5 5/8” long.

The work on the rails starts by taking the uncut blanks and edge gluing them together.  That’s done in two stages.  First is to glue up two pair of the blanks like this.  The F-clamps hold the blanks down flat to a base that has a plastic laminate surface so the glue does not stick to it.

Once the glue has cured the 2-piece blanks are glued together using the same process.  Since the glue joints are going to be cut away when the blank gets cut into strips, I don’t need a perfect joint and can use less glue.  Less glue means squeeze out is eliminated or drastically reduced translating into less of a problem if the already finished sanded surfaces need glue cleaned off.  

After the glue cures the clamps are removed and this is what the blank for the rails looks like.  It had just a tiny bit of glue squeeze out that gets removed with a sharp chisel. 

The rails need accurately spaced notches so when they go together the whole grid is square.  Rather than try and measure the notches out it’s more accurate to build a jig and use it to get constant spacing which again is more important here than the absolute distance.  The jig has a peg cut the same width as the saw kerf and is just a little shorter than the notch is deep.  Here it is attached to my sliding cutoff sled by a couple of clamps.  

This is a close-up view of the jig with the peg in place, the saw kerf and the blade which is adjusted to cut half way through the lattice rails.  Getting the right depth took a few tries using some scraps.  At this point the jig is ready to go.

The first pass is made 1” from the edge then that cut is set on the pin which lines up for a new cut 2” from the first.  That’s repeated 8 times giving this series of evenly space notches. 

Next is to rip off the rail blanks using the same setup and stop from earlier when all the lattice strips were ripped.  Each of the blanks that were glued together makes four rails plus a smidgen.  Here the fourth rail has been cut and you can see the sliver remaining to the left of the blade.  That gets cut off and the process repeats until the blank is completely gone giving me 16 rails.

As a check I did a test fit and the results were not what I had expected.  When assembled the rails were not flush top and bottom.  The left image shows what I wanted and the right what I got.

Surprised I rechecked my test pieces used to set the cut depth and they fit perfectly.  Next was to check the height of the pieces and both were the same.  For a bit I was at a loss as to what was going on but then it hit me.  First, the saw blade I am using to cut the notches does not leave a flat bottom like the cut on the left but leaves a little bit of a pyramid like the cut on the right.  I knew that and worked around it by cutting the notch just a little deeper.  The difference is the original test pieces I used to set the blade depth are about 1/32” bit thicker than the finished pieces but I was not concerned about the width of the cut just its depth.  So, to get them to fit together I had to make a second pass to widen the cut and therein lies the problem.  When I shifted the test pieces to widen the cut the blade removed most of the top of the pyramid making the notch seem deeper that it would in a single pass.  What that left me with was 16 rails with 9 notches each or 144 notches that are 1/64” too shallow.  The good or at least bearable news is that the problem can be fixed.  I just don’t know how long it will take.

Below is the setup and most of the tools used to fix the problem.  On the far-left side there is 1/8” spacer set in the end notch to keep everything aligned.  To its right is a clamp to help keep all the rails flat and in the same plane.  Next is a small thin fine toothed file and to its right is a Riffler.  Not shown is a razor knife I used to define the inside corner of the notch.  A CamClamp holds the pieces together and tight to a bench hook held in place with the vice.  As I got into the flow of cleaning up the notches each row probably took less than 5 minutes to do.

For those that don’t know this is one kind of a bench hook.

 

Next Up – Thicknessing the Rails/Lattice Pieces & Finishing the Jigs