Kumiko Sampler - #7 Making the Frames – Part 1

When the last post ended, I was trying to decide what to do with the stressed piece of mahogany that pinched the saw blade while it was being ripped.  Have to say that I am glad the splitter was in place.  Things could have gotten really exciting in an instant if it wasn’t installed.  When the partially ripped piece was set against a straight edge a bit of a curve on one side showed up.  Measuring each end of the board confirmed that the curved side had moved.  Here is a photo of the uncut end showing the board 3 5/8” wide.

This is the other end and it shows the board 3½” wide.  When I shut the saw down and removed the board the saw kerf had closed e.g. the pinch.  I tried to take another pass to open up the kerf and complete the cut but once again as the internal stresses were removed the kerf closed pinching the just cut edges against the splitter.  I started to try a third time but part way through the cut I decided I was pushing my luck and quit.  That’s why there is an open saw kerf at this end rather than you seeing the two sides of the kerf pinched together. 

After letting the piece sit overnight the kerf was still there and the board seemed to be stable with one edge still straight.  Based on that I decided to try and salvage the board.  The photo below shows the end result of three steps.  First, the edge that was opposite the one adjacent to the fence when the board pinched the fence is the straight one and it is at the top of the photo.  That straight edge was put against the fence and run through the saw truing up the curved edge giving me straight parallel edges.  Second, the fence is reset a bit wide of the previously ripped blanks and with the original straight edge against the fence a blank is cut.  Third, the board is flipped so the new straight true edge is against the fence and it is ripped until the cut meets the original pinched cut giving me a slightly wide blank.  I have to say that I was a little surprised that it cut just fine with no pinching at all.  It seems all the stress was released in the first set of cuts.  The offset cut in the middle piece shows how much the board moved as the stresses were relieved.  Resetting the fence to match the other blanks and ripping gives me two usable blanks. 

The last step is to stand the blanks on edge and pass them through the thickness sander to remove any surface irregularities and smooth out the saw cut.

In looking at the frame profile I have decided to revise the plans again.  The top photo below is the previous version and the bottom one is what I am thinking about now.  It allows for the frame to be finished before the plexiglass is installed and I think will make for a simpler assembly of the glass, sampler, matt board and foam core backer.  Although it may change some more before I get done.

Next is to decide then mark which face will be out and the top of each of the blanks.  The idea is to align the blanks so that any flaws are either cut away or hidden from view when the frame is completed.  There are also three pieces to be used for testing.

Cutting the dado in the frame is next.  It starts with a ¼” stacked dado blade almost centered on the blank set to cut just shy of ¼” deep.  Here is what the setup and first pass look like. 

For the second pass the fence is moved over about ¼” and the blank run through again.

The blank is flipped end for end and run through the saw again.

A third pass is made along the top edge leaving an uncut strip slightly over ¼” wide.  That’s followed by another pass at the bottom leaving a 3/16” wide strip.  In the photo below the thin edge is at the top.  That edge is narrower to give the needed depth in the frame for the glass, sampler, black matt board and a foam core backer.  If the blanks had been cut a little wider then that would not have had to be done but since I changed the design, I needed to pick up just a little more depth.  Also, as you can see the bottom of the dado is not the greatest looking.  If this were a typical frame that would not be a problem as it would all be hidden.  However, in this case because the sampler is about half an inch tall and the inside of the frame will be visible it needs to be cleaned up.

The cleanup is done using a straight flat end router bit taking off just enough to clean up and flatten the bottom of the dado.  Here you can see the bit and its relation to the frame blank. 

After cleaning up the dado’s bottom a small radius is added to the top corners of the blank using the router.  This photo also gives a good look at the narrower leg that I had to make to give me a little more frame depth.

Removing the thin leg is next and is done on the table saw.  The blank on the left up next to the fence is ready to cut while the one on the right has had the leg removed.  The thin strip is all that was cut off.

When I used the router to clean up the bottom of the dado.  I did not want to trim the inside face of the leg and in doing so left a little strip that needs to be cleaned up.  It’s really small but you can just see it circled in red at either side of the dado.

It’s easy to cleanup by gently using a scraper and then some light sanding.  At this point the blanks are ready to have their ends mitered and cut to length.

Next Up – Completing the Frame, Lacquering & Assembly

Kumiko Sampler - #6 Completing Lattice Infill & Starting on Frame

With the jigs set for making the wings the beveling and gluing in place of them can begin.  After some experimentation I found a good work flow for cutting and installation.  It starts with cutting four wings then doing a test fit.  In the photo below that would be the right-center top space.  There are also four more cut wings at the bottom.  So far each of the test fits has gone well with no additional work necessary.  After the test fit a dab of glue is added in the corners and the four wings are set in place.  I put one finger on one pair of the wings meeting points, my thumb on the other pair and squeeze them together for one minute.  That’s enough time for the glue to grab and hold.  I have to say the cutting and gluing of the wings in place is taking me less time than I had expected.  Also, only cutting four wings at a time then gluing them in gives my fingers a break.

Once the multitude of wings are installed, I am left with only the kingpin group to cut and install.  As the pattern has progressed the pieces have gotten smaller.  Now the blanks for these last pieces are less than ¾” long and once cut to length they have to have a pair of bevels cut on each end.  Here is what the blank looks like set in the jig.  Note the hold down set just to the right and above the blank.   In use the left end sets on the blank holding it in place allowing my fingers to be away from the razor-sharp chisel.

A few passes with the chisel paring off thin shavings brings the edge down flush with the face of the jig.  Looking close at the enlarged view you can see where the bevel does not go all the way to the bottom of the piece.  That’s because when it gets flipped over and the other side beveled the intersection of the two bevels will meet in the center of the piece.

Here with one face done and the blank flipped over the second side is ready to be pared off.  After that the blank is flipped end for end the process is repeated to make the finished kingpin.

Here is a closeup of the kingpin.

Now it’s just a matter of cutting and gluing in all 16 of the kingpins.  When done this is what the glued-up sampler looks like.  It’s still unfinished as not all the pieces are absolutely flat and in the same plane.  Some are a little high and some a little low.

To flatten the sampler 120 grit sandpaper is clamped down to a flat surface and the completed sampler is scrubbed back and forth rotating it to get a flat and even surface.  Once flat a three-step set of sandpapers is clamped down with 220 on the left, 320 in the center and 400 on the right.  The sampler gets the finish face worked through each grit ending with the 400.

Below is the sampler after it has been through all the grits and is finish sanded.  It may be a little hard to see in the photo but in person you can certainly tell the difference prior to flattening and the finished product.

The half sampler had different length tails on the different sides so I decided to cut them flush with the perimeter rails.  The Japanese pull saw is used to make the cut a fraction of an inch long.  The saw was guided and kept square by a piece of plywood that hung over the rail all on top of a piece of Styrofoam that protected my bench from being cut by the saw.

To remove the little bit of stub and bring the tail flush with the rail some 120-grit sandpaper along with a block to keep the sample plumb make quick work of smoothing things out.  Once flush the 120-grit sandpaper was switched out to 220.  I think I may band this sampler or something else but am not sure yet.

What I am planning on is to mount the full sampler in a frame with a piece of plexiglass for protection and to keep the dust out.  This is what I have in mind for the overall design.  Of course, it’s subject to revision whenever I feel like it.  That’s the advantage of being the owner, architect and builder. 😊

However as always, the devil is in the details.  So far, I am looking at two different frame details.  In the drawings below the top one has a space between the sampler and the glass while the bottom one does not.  Of the two I am leaning toward the top one because I think that not having the sampler right up against the glass may work better.

My original thought was to make the frame out of cherry but that was when I was only making a single sampler and now there are four.  While going through my wood stock looking for the right pieces of cherry, I came across a couple pieces of mahogany.   They were from a friend who when his eyesight failed him he quit woodworking and gifted them on to me.  He had made a tall hall clock with the material and it is truly beautiful.  So now one frame will be of cherry (top board) and the other out of mahogany (bottom board).

While the cherry piece was straight and true the mahogany had only rough edges to work with plus a little bend in it.  A trip to the bandsaw gave me a pretty straight albeit rough edge to work from.

A few passes using the long plane cleaned up the edge so I had a good reference surface to work with.  Next was to check the boards for flatness and two of them had a bit of twist.  A little work with the hand plane and running them through the thickness sander got them flat and all to the same thickness.

A trip to the table saw to rip the boards into the blanks is next.  The cherry and one of the pieces of mahogany cut down fine.  However, the other mahogany board had some internal stresses that as I made the cut were released pinching the blade some even with the splitter in place.  Here is as far as I could get in the cut before things went bad.  The splitter is the black curved piece of steel just back of the blade.  Now my options are to try and salvage the board or cut another piece of mahogany to replace this one and start over.  However that's a decision for tomorrow.

Next Up – Making the Frames – Part 1

Kumiko Sampler - #5 Chisel Sharpening & Lattice Infill – Part 1

Before starting on the infill pieces the chisel that will be used to cut the bevels on the end of each piece needs to be sharpened.  It’s been quite a while since it has had any serious sharpening done other than some honing on a leather strop.  Over time it has gotten a few small nicks in the cutting edge.  To get rid of those I decided to go through a full sharpening series. 

I sharpen my chisels and hand plane blades using a jig that holds the tool at the correct angle.  Some people are good enough to do that by feel but not me.  The photo below shows the jig in use.  On the left side of the photo is a black gauge with red, yellow and green writing.  It attaches to the jig and is used to set the sharpening angle then removed.  For the actual sharpening I have used natural stones, Japanese water stones, diamond plates and what’s used here.  It’s wet/dry silicon carbide sandpaper on a flat piece of granite using water as a lubricant.  I like this method since I can get the paper in grits ranging from 180 up to 2,000 plus they are fairly inexpensive and the granite always stays flat.

When the final sharpening pass is done with the 2,000 grit the last step is to polish the edge using a leather strop.  At this point the sharpened edge is near a mirror finish.

Now the actual cutting and fitting all of the infill pieces can begin.  First are the eight diagonals.  They are about 1 5/8” long with a 45-degree bevel on both sides of each end.  This drawing shows what it will look like when done. 

The first piece is cut a fraction long using a stop block clamped to the bench hook and cut with the Japanese pull saw. 

Next using the 45-degree angle jig the length stop is set to cut a bevel just ½ the thickness of the piece.  Here the unbeveled piece is set in the jig and is ready to be cut.  A finished bevel requires three or four passes taking off a little at a time until the chisel rides flat on the angled jig face.

With one side beveled the piece is flipped over and the other side done.  The piece is then flipped end for end and bevels are cut on the other end.  Here is a closer view of a completed pair of bevels.

The completed diagonal is then test fit.  Since this is the first try to no surprise it was a little long.  Going back to the beveling jig the stop is adjusted and the bevels on one end recut then back for a test fit.  The process is repeated until the piece just fits.  Too tight and the grid of rails will be thrown out of square, too loose and you end up with a gap.  The margin for error is really small.  When a good fit is achieved the overall cut length of the blank is adjusted on the bench hook and the diagonal blanks cut.  Here are five of the eight diagonals installed. 

When everything works out right this is what one of the squares looks like where all the pieces come together.  After all the diagonals are cut, they are removed then glued into place.

After the diagonals comes the wings and there are 32 of them.  Here is a drawing of what the sampler looks like when they are added. 

Like the diagonals they are beveled on each end.  Unlike them the angles on each end are different and on one end they are not centered.

The wings begin by measuring and cutting a couple of blanks to the approximate length then using the 22.5 and 67.5 degree jigs bevel the ends and do a test fit.  As the blanks were long the same process of making adjustments and fitting that was done for the diagonals is done for the wings.  With the wings really, really close but long I could set the stop on the bench hook and start cutting the blanks needed.  The 90 blanks I cut are shown at the top and the almost fit wings are in place in the lower left of the sampler below. 

Why, you ask, are there 90 blanks when only 32 are needed for the sampler?  Well, that’s because in a moment of weakness I decided to make a second full sampler, a half sampler and one small ¼ sampler. 

The process for the final fitting of the wings is to cut the 22.5-degree double bevel on one end of all 90 of the blanks.  That has to be done in phases as after about 45 minutes my fingers are ready to take a break.  This is the setup for doing those bevels.  The other jig has a 67.5-degree bevel and is used for the other end that I used to do the test fitting.  I can’t cut both ends in production because the 67.5-degree end is an unequal split which needs two 67.5-degree jigs using different length stops.  Don’t worry if that does not make sense as it will get covered in a bit.  In the photo in the upper right you can see the leather strop used to touch up the chisel edge every 5 pieces or so.

This is a little closer view of bevel setup.  One minor correction is that most of the time a skew or angled cut is used to pare off the wood and make the bevel rather than the square cut shown here.

Once the 22.5-degree double bevel has been cut on the wings the setup and fitting can begin for the two 67.5-degree bevels on the other end.  As noted above the bevels at this end do not meet in the center but are offset about one-third, two-thirds.  That all has to do with the way the wings and kingpin meet.  Below in the drawing and blowup of the red circled area you can see what I mean.

In order to get the unequal, split the stops on the two 67.5-degree jigs are set at slightly different lengths.  One jig has a depth stop already set close for the two-third bevel from earlier on when doing some rough testing to determine the wing blank length.  The second 67.5-degree jig is set a hair shorter for the one-third bevel.  As before trial and error dials in the right combination so the 22.5-degree double bevel fits tight in the corner.  Here are the two jigs set up and ready to go.  In case you are interested the second cut for the one-third bevel stop is set .02” shorter than the two-thirds one.

Next Up – Completing Lattice Infill & Starting on Frame

Kumiko Sampler - #4 Thicknessing the Lattice/Rails & Finishing the Jigs

With the notches having their bottom flattened the next step is to thickness sand all of the pieces down to 1/8” thick.  When ripped on the table saw they were cut just about 1/32” too thick to allow for any saw marks, unevenness in the cut and to sand down to a nice smooth finished surface.  Here are all the pieces laid out in five groups ready to be sanded.

The top group of lattices have been run through the sander once and are much smoother.  It will take several passes to achieve the desired thickness.  Not because there is a lot of material to remove but because I don’t want to take too much off and end up with a bunch of scrap.  There is no going back if I take too much off.   The bottom group of rails are on a carrier.  It helps me keep all the pieces together and easy to feed through the sander.  They have not been run through the sander yet and still have the diagonal pencil lines on them.  The lines give me a good visual regarding the status of the surface as the passes through the sander are made.  When they are gone the surface is flat and has been completely sanded.

It did take several passes making tiny adjustments right at the end to get all the pieces to the desired thickness.  With that done I set the pieces aside and went back to work on the jigs.  There are two pieces yet to make, a stop for setting the piece length and a hold down to keep the pieces in place while the angles are being cut.  In the drawing below they are the two pieces with dimensions. 

I started with the stops which requires a slot cut in the center where the locking screw will go.  It starts by drilling holes to establish either end of the slot.  A small pilot hole is drilled then bigger bits are used to enlarge the hole.  Here the hole on the far-right has been drilled out with a larger bit.  I started with a small bit because I was worried that using the finish size bit might wander and not give me a centered hole.  A couple of steps in bit sizes kept the hole centered where it needed to be.  There are four holes because both stops are being made from a single blank that will be cut apart later.

Once the holes are drilled the piece is clamped in the vice, the coping saw blade is fed through the just drilled end hole and the slot is cut.  My preference for this type of work is to install the blade with the teeth pointing toward me so it cut on the pull or in this case the down stroke. 

Once cut the inside faces are smoothed using rifflers and a file.  Here the blank has been cut in half and installed in the jig.  There are three holes in the jig so the stop can be adjusted to whatever length lattice piece is needed.  The big hole on the left at the bottom of the dado is from when this piece was used as a mantel support. 

The last part is the hold down.  Like the stop both are made at the same time cut out of a single piece.  Once ripped to height and width the notch in the base is laid out and rough cut on the bandsaw.  Finish cutting is done on the router in this setup.  If you are wondering how I routed the square corners at the end of the cut I didn’t.  It shows that way because I forgot to take a photo during the routing and didn’t remember until I had already squared them off with a chisel.  From here they are cut to length on the chop saw. 

With the angle jigs along with the lattice and rail blanks completed I believe there is just one last item that needs to be made before the assembly can start.  The components for the Kumiko sampler be it rails, diagonals, wings or kingpins all need to be cut to length.  Since everything but the rails are less than 2” long with most of them being shorter and just under an eighth of an inch thick I am a little, no make that a lot leery of cutting them on the chop or table saw.  At least to start out the plans are to use a Japanese pull saw along with a modified bench hook and stop to make the cuts.  The Japanese pull saw’s blade is only about 1/100 of an inch thick which will minimize the kerf waste when making all the cuts.  Modifying the bench hook consists of cutting a plumb and square kerf in the backstop using the pull saw.  A couple of thick oak blocks clamped to the backstop pressed tight to the saw keeps it square and plumb while I make the cut.

The three long rails get cut first followed by the five short ones.  They are cut in sets using a scrap piece of rail as a key to keep them in alignment and a clamp to hold them in place. 

A test assembly is next which lets me know if any of the joints are too tight or too loose.  As it turned out two of the joints were just a little tight.  A little sanding and they fit just right. 

Glue-up is next with one crosspiece left in place to help with alignment.  A tiny drop of glue is applied to the inner faces of the notch on the long rail using the thin pointed applicator in the lower left.  When done any glue squeeze out is removed with a small snap blade utility knife.  No need for an after photo as it looks just like the test fit photo above. 

Next Up - Chisel Sharpening & Lattice Infill - Part 1