Kumiko Sampler - #2 Lattice Blanks & More Jig Work

In the Fine Homebuilding article, the author suggests using basswood or pine as both are fairly soft and easy to work.  I believe the author is on the east coast and as such think he is referencing white pine.  In thinking about what to use that’s on hand I considered cedar, alder and radiata pine.  Both the cedar and the pine were eliminated leaving the alder because I am not fond of doing fine work with coniferous woods.  I don’t like the pitch they leave on tools or how they gum up sandpaper and the plans are to use the thickness sander to get all the lattice work pieces to final width and thickness.

The alder plank is a little over 6” wide and about 6’ long.  One end has a lot of imperfections like knots, cracks and bark inclusions.  Normally that’s a problem when trying to get clear material but the lattice pieces are only ½” wide x 1/8” thick.  Length varies from 1 to 8 inches so I think it will be pretty easy to work around the flaws.  Here I have cut 18” off the plank and ripped it in half giving me two pieces a little over an inch thick 18” long and 3” wide.

To get my ½” wide pieces I will rip the 18” long pieces down the 1”+ width.  Using a marking gauge gives me my cut line or should I say cut lines.  I could spend time measuring and testing to make sure the line is exactly in the center of the piece but I find it easier to get close and mark from both sides.  The result is a pair of lines with the center between them.  If you look closely you can see them in the photo below.  Now all I have to do is keep the bandsaw blade between them.

Now it’s off to the bandsaw to rip them in half and in doing so find out how bad the corrugated steel fasteners damaged the blade. 

After ripping both pieces, the blade is not completely ruined but I would say that its lost probably 70% of it useful life.  Alder is pretty soft so when I go to rip something like oak or cherry the useful life may go down even more.  Looks like I am going to be ordering a replacement blade.  Here’s the ripped faces of the two pieces.

While the ripped lattice pieces rest to see if they are stable it’s back to doing some work on the jigs.  This is my dimensioned drawing of them.

Since lattice pieces are ½” wide the first step is to cut a dado that wide and ¼” deep.  It can be cut with either a router or the table saw.  I decided to use a stacked dado blade on the table saw and get it done in a single pass.

To make the lattice piece drop in the dado a little easier a block plane is use to put a small bevel on the dado’s edges.  Also, the top and bottom edges are rounded.   That could have been done with the router but it’s a lot faster to use a hand plane.  Long before the bit could have been changed and adjusted, I was done using the radius cutting hand plane.

Cutting the jig’s end angles are next.  Three of the four are easily done by setting the 45 or 67.5 degree angle on the saw and making the cut.  The last cut is 22.5 degrees which requires the jig’s narrow end set up next to the saw’s fence.  Making that type of cut is not safe so it requires a little bit of help to securely hold the jig in place while the cut is being made.  The photo below shows the mix of clamps, spacers and blocks needed.

Here are the bodies of the two jigs almost done.  There are still some parts to be made but that will come later.  Tomorrow I can move back to working on the lattice blanks.

Next is to check the lattice blanks to see if they are still flat and true.  Fortunately, they were so I could move on to cutting them to the same width using the table saw.  After that the blanks are run through the thickness sander until they just drop into the ½” wide dado in the jigs.  That means they are about 1/100” shy of ½”.  Here it’s more important for all the blanks to be the same thickness rather than an exact thickness.  Below are the four blanks at their finish thickness and since the sandpaper in the sander is 220 grit they are also finish sanded.

Ripping the ½” blanks down into individual pieces just over 1/8” thick is next.  That’s done on the table saw and which blade to use is the first decision.  A specialized rip blade is used but I have two different thickness.  One cuts a 1/8” thick kerf and the other is a thin kerf blade that cuts a 3/32” kerf.  That may not sound like much but the 1/8” blade is about a third thicker and that adds up.  Most of the time I use the 1/8” blade.  However, in this case that would turn right at 50% of the blank into sawdust.  Using a thinner blade will give me one extra lattice strip per blank.  Here is what the two blades look like side by side.

There are at least two ways to rip the blank down to the required width using the table saw.  First is to set the fence to the desired width and rip away.  That works great if the needed piece is wide enough to do safely but with roughly 1/8” wide pieces that’s too narrow for me even using a push block.  The alternate is to keep what you cut off the blank and reset the fence for each cut.  That can be a problem in trying to get identical pieces.  However, there is a work around for that.  A simple jig allows the fence to be quickly and accurately reset.  Here is what the jig looks like setup and ready to go.  In use all four blanks have a piece cut off giving four lattice pieces per setup.  After all four blanks have had a piece ripped off the blank is set against the jig, the fence pulled up snug to the blank, locked in place and that’s it.  Cut four more lattice pieces and repeat and repeat and repeat.

Backtracking just a little to show how the jig is set here you can see the dial caliper set at 5/32”.  That will give me a 1/32” thicker lattice piece than I need.  Running them through the thickness sander will let me remove the saw marks and any slight variations in thickness ending up at my 1/8” thickness.

Next Up – Making Slats & Rails

Kumiko Sampler - #1 The Beginning

I have subscribed to Fine Woodworking for a long time and in their February 2017 issue they had an article on a Japanese form of latticework called Kumiko.  After reading the article it sounded like something that I would like to try sometime in the future so I set the issue aside in my “maybe-to-do” stack.

Well, with the Covid-19 restrictions in place I found myself in self-enforced seclusion here in the desert.  Normally we take some time off to visit a cooler climate during the hottest part of the summer but that was not going to happen this year.  As the shop cools off at night and is comfortable for a few hours in the morning I decided to try and see if I could make a small section of the Kumiko latticework called asa-no-ha.

First off, as usual, I drew up a 3D model in SketchUp which looks like this.  The lattice itself is about 4” by 8”.

With the model done I wondered what I could do with the piece when it was finished.  Putting it in a cabinet as a decorative inset came to mind but we really don’t need another cabinet.  Then I remembered a 3D optical illusion piece my dad had done several years ago and given to us as a gift. 

Taking a cue from that it was back to the 3D model to put the lattice in a small self-contained display case.  This is what that looks like using a black matt board background and cherry as a frame.  Right now, it’s designed as a roughly 7” by 11” wall hanging.  Now that might change as the project progresses.

There are a couple of jigs the build will require so I decided to start with those.  Recently I was gifted an oak fireplace mantel along with its supports.  The supports had mounting holes in them along with some face glued pieces that were also nailed on so they were not pristine pieces but would certainly be usable for the jigs.  Below is one of them to give an idea of what I started with.  To locate the nails for removal I used a rare earth magnet (circled in red) marking each one as I went.  The nails had been set and filled which made their removal a little more challenging.   One thing this assembly clearly shows off is what happens when wood is glued up with the grain running at 90 degrees to each other.  All of the face glued pieces were split as clearly shown here.  Wood does not change much at all when the humidity changes along the grain but it can change a lot across the grain.  When you have wide piece like this glued and nailed to another piece where the grain does not align then somethings got to give.  In this case the thinner piece of wood split to release the stress. 

In order to get to the nails, I put a plug cutter in the drill press then drilled a plug centered on each nail.  Here you can see the setup, some cut plugs and the remaining nails to be cut circled in pencil.

Once the plugs are drilled a cold chisel is used to split the plug from the nail.   After the waste is removed a pair of wire cutters are used to pull the nails.  I used a cold chisel rather than one of my good wood chisels because if one of the nails were hit it would have damaged the wood chisel’s edge.

Next is to cut face glued pieces flush with the underlying 2” thick oak.  That’s done on the bandsaw and the new blade made quick work just gliding though 1 ½” of old red oak like it was nothing.

Cutting the legs free from each other along their joint is next.  After a careful check for any screws or nails I used the bandsaw to cut the joint.  Unfortunately, when I looked at the just made cut, I saw two shiny pieces of freshly cut steel and that’s never a good sign in woodworking.   

At that time, I no idea what had been hit only that my new bandsaw blade wasn’t new anymore.  After splitting the face glued oak off the second oak support I found out what I had hit.  It was a pair of corrugated steel fasteners, one on each side of the 2” thick oak.  They were not visible when looking at the face glued pieces and were too deep in the wood to be found with the magnet.  All that’s left now is to ascertain how much damage was done to the new bandsaw blade.

I had planned on cutting off the face glued pieces with the bandsaw but was a little paranoid if any other surprises were hiding.  Splitting off the face glued pieces resolved that but sure did leave a surface in poor shape as in the top photo.  A little work with the hand plane gives me a flat surface to work from.

Here are all four pieces flattened and ready to be run through the thickness sander.  The notch on the left end of the upper two pieces are where the corrugated steel fasteners have been cut away.

The photo below shows the two pieces I will use for the jigs after they were run through the thickness sander, ripped to give clean square edges and have had the ends trimmed on the chop saw.  The lines on the jigs are marked for the different angles needed for cutting the ends of the lattice pieces, 45, 67.5 and 22.5 degrees.  Fortunately, I have a steel protractor that has marking slots at all the needed angles including 67.5 and 22.5 degree. 

At this point I am going to set the jigs aside for a bit to see if the wood is stable and start on the lattice material.

Next Up – Prepping Lattice Blanks & More Jig Work  

Wood Body Pinhole Camera - #11 Photos & A Useless Item

I had hoped to be able to add some photos taken by the camera this week but the developed images have not been received yet.  When they do I will add them to this post.

UPDATE: July 9, 2020

Rob sent me some photos from the first roll of film through the camera and here are four of them.  While they are not modern digital crystal clear images they certainly convey the character and ambiance of early pinhole photographs which, after all, was the intent of the camera.

The useless item is a laminated hardwood ball made out of leftover walnut from the camera and some thin maple leftover from I don’t know what.  In this photo of the initial material selection not all the two pieces of walnut with the red arrows were used. 

As the goal is to make as big a ball as possible using just these materials some measuring and planning on paper is needed.  Here is everything I used to decide the layout and lamination order. 

Once cut to size here is the order the pieces will get glued up to make the turning blank.

I started from the center and worked outward.  The maple pieces are first and they get glued to the center walnut block.  To keep them from sliding around after the glue is applied and while being clamped up a couple of small screws are used to assure correct registration.   Once the glue cures the screws will be removed and as you can see, they are outside the ball limits so will be removed. 

 A couple of big clamps apply pressure and are left on overnight to allow the glue to cure.

The next day the clamps and the screws are removed.  Now the outer two pieces of walnut can get glued on.  I didn’t use screws to keep them in place but instead use some large C clamps to hold the edges flush while the large bar clamps are tightened.  The photo below shows just the large bar clamps in place.  Note the triangle on the stack.  That gives me a quick visual clue that all the pieces are glued up in the right order and facing the right direction.  I know that sounds a little funny but because the pieces are square it’s easy to get one turned 90 degrees from where it should be.  That’s the voice of experience talking.  Fortunately, I found that mistake before the glue had set and was able to correct the problem. 

Another overnight cure and a little trim of the piece gives me not a cube but a square piece taller than it is wide.  Since this will be turned between centers on the lathe the blank needs to be taller than the balls finished diameter.  Don’t worry if that’s not clear later on you will see what I mean.  The photo on the left shows the blank ready to be cut on the bandsaw and the photo on the right is after it’s been cut roughly round.  I do that as it’s a lot easier to knock off the corners with the bandsaw rather than using the lathe.

Here is the blank after it has been turned round and to a constant diameter.  Everything outside the white pencil lines gets removed in the first step of turning.  They are really just for illustration because as soon as the turning starts they will go away. 

The turning starts out by eyeballing the arc of material that gets removed but as I get closer a template is used to make sure I have the right arc.  This template is cut from a sheet of typing paper that had an early version of the camera front.  Here the initial turning is done and the center line of the ball marked in white.  You can see the stubs from the turning points which is why the blank starts as a tall square and not a true cube.

Next is to remove the drive and tail centers replacing them with some wood turning jigs I made for making balls.  The partially completed ball is set into them turned 90 degrees so the knobs left from the step 1 turning can be turned away.  

When step 2 is done, I will have a piece with an X and Y axis trued up.  Here is what that looks like with  two axes marked in white turned.  Side note, did you know that the plural of “axis” is “axes”.  Yes, axes is the plural of ax, axe and axis making it the distinction of being the only word in English that is the plural of three different singular nouns.

Step 3 is to rotate the piece again so the Y axis can be trued.  The goal is to cut the piece so on each axis it spins true without removing the center line from the previous setup.  Cutting away that centerline will reduce the diameter on that axis change a sphere to an oblong.  It can take rotating through the steps a couple of times to get a good sphere.  When the turning is done the sphere is sanded rotating through all three axes.  Here all three are done with each marked in white. 

Last is to apply a finish.  Since this is not going to have a hard life or get used very much, I decided to use a wax-based friction finish.  To apply the wax stick is held against the rapidly turning sphere until the friction melts some onto the surface.  It is then buffed out using a soft cloth again held against the spinning sphere.  Here is the setup with two pieces of an old sock set between the turning jigs that act as a cushion to prevent marring of the finish.  To get an even finish the process needs to be repeated on each of the three axes twice.  Buffing the sphere is a balance between applying enough pressure to generate enough heat to melt the wax but not applying so much that the friction in buffing exceeds the friction between the sphere and the sock pads.  If that happens then the sphere stops spinning while the drive jig keeps spinning marring the finish on that end.  That’s fixed by rotating the sphere and rebuffing that area.

This is the finished product temporally set on one of the jigs  ready to take its place as a purely decorative item.