Spindles

I thought that I had used up all my mistakes earlier.  Today I was proved wrong.

I decided to go with 1-3/8” spindles, but my 8/4 stock measured in at about 1-7/8”.  I could have simply planed it down to the right thickness, but that seemed like a huge waste of some very expensive wood, so I figured I could resaw it to 1-1/2” than plane from there.  This would leave me with a thin slice of “waste” that I can recycle into other projects.  To start, I had to change the blade on my band saw back to the 3/4" for the resaw.  I did this fairly quickly, but something was just not right.  The blade tended to “flutter”.  I tweaked the tension, the guides, and the tracking, but no joy.  I just couldn’t get it to run smoothly.  Hey, it was going around, so I went ahead and started cutting.   As I got to the last two feet of the 6’ board, I noticed some smoke, but assumed the blade was just cooking a bit because the wood had some internal stress and was squeezing against it.  Then the blade took a hard right from the cutting line.  I tried to get back on line, but it was a struggle.  When I finally made it completely through the board, I saw that although the blade only moved a little at the top, it moved significantly on the bottom.  In fact enough to go beyond the 1-3/8” I was shooting for.  Arrrgh!  That end of the board was wasted, so I would have to use another 8/4 board (my last) to make the rest of the spindles.  I took a look around the band saw and found the culprit.  I managed to burn out the bottom rear guide bearing.  The good news is that this happened before and I bought a spare bearing.

After replacing the bearing, and switching back to my new 3/8” blade, I marked out the pattern and cut all of the spindle side profiles.  Wow, this is a lot of cutting!  Next, I traced out the front profile on the first spindle.  Following Brock’s recommendation, I also marked the center of the piece.  I cut the right side with no problem, but when I went to cut the left side, I inadvertently started cutting the centerline.  Doh!  Good thing I cut two extra spindles.  I’ll use this one as a practice piece to test the shaping and the fit to the chair.  Lee Valley shipped the dowel cutters (a.k.a. pencil sharpeners) today.

Lesson Learned:  If you are going to draw lines on the stock that will NOT be cut, use a different colored pencil.

I finished cutting out the rest of the spindles and they came out pretty good.  The radii on the bottom of the spindles were too tight to cut on the band saw, so I cut those using my scroll saw.  Notice that I renumbered the #9 spindle to be #1B.  :)

I was limited in my choices as far as grain selection.  There may be some “short grain” issues that would weaken the spindles, but I guess I’ll find that out once they are in the chair.  I think the current batch matches well.  If I have to use #8, there will be some lighter heartwood showing.  One of the spindles had a knot in it and I was worried about that, but I cut it out anyway.  It turns out that the knot is now just grazing the corner of the upper spindle and that part will be shaped away.

I’ll have to get more 8/4 walnut before I can make the rockers.  While I’m waiting on that, I think I’ll move on to some shaping and get the legs, arms, and headrest screwed into place.  I’m finding that at least 50% of my time goes into tool setup, planning, watching the DVD, or just staring at Maloof chair pictures and trying to figure out the lines.  I’m frustrated with my mistakes again, but I have made progress.  As long as things are moving forward, I know I’ll reach the end eventually.

Arms

Christmas was wonderful.  I’m a morning person, so while the rest of the family was sleeping in, I was busy reviewing the Brock DVD and book, or sharpening tools.  I don’t really like to sharpen things, but I know it’s necessary.  So, I try to dedicate a day every quarter to just sharpen every tool I have: knives, planes, chisels, turning tools, etc. 

Santa was good to me, so I went shopping at Woodcraft today and picked up a few more tools.  I picked up the Miller dowel 1X drill bit, the “Dowel-it” jig, and some Titebond III long open time glue for the rockers.  I also stopped by Harbor Freight to pick up ten 6” C-clamps (on sale!), and Home Depot (a.k.a. BORG…Big Orange Retail Giant) for some Spax #10x3” screws.

In my “free time”, I came across the Sawmill Creek woodworking forum.  There is a great discussion thread there on Maloof rockers with 5 or 6 people who are also currently building chairs.  It was great to see that other people are experiencing some of the same issues I’m having.   I also learned a lot by reading the problems and solutions of others.  That’s what I hope this blog will do for the readers. 

I was struggling with my first arm.  The plans call for 6 degree cuts on the bottom/front of the stock where it attaches to the front leg, and the back where the arm attaches to the back leg.  I quickly realized that because of my earlier mistake, the arm transition on the back legs were now 1” higher than planned, so the 6 degree elevation was not working.  After some trial and error, I found that 8 degrees seemed to work.  Once things were aligned, I traced the spot where the front leg contacts the arm stock.  I recommend using stock that is slightly longer than required (~20”).  This will give you more material to play when working on the fit.

The plans call for a 1/2" dowel center to be placed in the top of the arm, and then the arm stock is pounded down to mark the spot to drill for the dowel.  I don’t have a dowel center, so I simply used my compass to find the center of the circle I had drawn.  Then I used my new dowel jig and a 1/2" brad point bit to make a 1-1/4” deep hole.

Next, I traced the top and side arm templates to the stock.  You have to be careful of the orientation of the top template to ensure you cut a right and a left arm.  I marked the template on the bottom of the stock, that way I can ensure that the inside edge is 1/2" from the leg circle I traced.  In the Brock DVD, he cuts one side of the arm, and then the other by carefully balancing the arm.  This seems difficult to me.  I use the technique I learned from years of compound cutting on the scroll saw. I start by cutting the top and bottom areas off, and then I use clear packing tape to reassemble the parts.  This method provides a flat surface for cutting the other face.  Here are some pictures to show you what I’m talking about.

I created some dowels for the front leg joint and mounted the arms to the chair.  For some reason the arms fit even better than they did before I cut them to shape!  I should mention that I created a simple jig to hold the back of the arm in place.  This was a big help when trying to get the arm-to-back leg cuts to match.  Now the chair is really starting to look like a chair.  My goal this week is to complete the rough cut of all the parts.  Later, I can take my time to shape everything.

I was very interested in the Miller dowel bit and Spax screw combination.  I did a test using a scrap piece of wood and it worked well.  However, I wanted to see exactly how it worked.  So, I cut through the wood to take a look inside.  This picture shows that the shank of the screw is not touching the wood, but the threads on the end are.  This is due to the stepped shape of the bit.  When screwing together two pieces of wood, this will result in the inside piece being pulled back into the outer piece for a very strong joint.  The Miller bit also does the countersinking to prepare for the plugs.

Headrest (Part 2)

Today was all about the headrest.  There were no patterns provided for the headrest shape.  This allows me to be creative, but it also means I need to decide on a design and create my own pattern.  The Brock chair uses a nice looking French curve, but I decided to keep it simple with a shallow curve on the top, and a more pronounced curve on the bottom that flows into the legs.  This closely matches the Maloof chair.  Using a thin strip of maple, I created an adjustable curve to trace the arc.  Because the headrest is now curved, I drew the curve on some poster board so it can be bent on the wood.  The curves into the legs were just drawn free hand.  I cut out one side, and then folded it over to trace and cut the other side.  This ensures that the curve is symmetrical.  I suggest that you first cut the curves out on some plywood.  That way if the poster board pattern gets worn out or torn, you can quickly create another one.

Here is a picture of the curves traced onto the headrest.  I mounted it onto the chair so the curves could be drawn on the legs too.  The Brock video shows him cutting the curved headrest with no support.  I saved the back of the stock to use as a “cradle” to hold the headrest stable and make the cuts straight.

It really helps to cut just the BOTTOM curves first.  Then the holes for the spindles can be drill easier because the top is still flat.  The holes are 3/8” x 1-1/4” deep, and 2-7/8" apart.  They are set back 7/8" from the front.  After drilling, the top curve can be cut.  Here’s a picture of the finished product.

Back Legs (Part 3) + Headrest

Subtitle:  Why didn’t I just stay in bed?

My motivation for this blog was to share my mistakes so that anyone else building a chair could learn from them.  Well, prepare for a huge learning experience!

This post covers the last two days, since I didn’t get much done yesterday.  I ordered a new band saw blade From Grizzly on Monday morning and it arrived Tuesday afternoon!  Of course, Grizzly is based just an hour and a half north of me in Bellingham, WA, but I was still impressed with the response time!  Thanks Griz.  You guys earned a link from me.

The blade I received is a 105” x 3/8”, 4 TPI, Timber Wolf brand.  This thing breezes through 3” walnut like butter.  No burning, no squealing.  I wish I had bought this a long time ago.

While I was still waiting on the blade, I cut the adder blocks on the back legs using my table saw from hell.  I clamped it to the 6 degree jig to get the correct angle.  As you can see from the picture, the clamps are dangerously close to the blade, so I had to be very careful.  Due to the opposite angles on each leg, the 6 degree jig has to be flipped 180 degrees for the second leg.  The only problem I experienced was because I didn’t check that the blade was square with the saw table.  I recently made some boxes that required the blade to be tilted at 45 degrees.  When I moved the blade back, I only used the markings on the saw.  I neglected to check the blade with a square.  When the blade is low, this isn’t much of a problem, but when it’s up at its full height (as it was here for the adder blocks) the difference is clearly visible.  This gave me some more practice with my hand plane to ensure the cut face of the adder block was square with the flat side of the leg.

Lesson learned:  ALWAYS check the square of your blades, especially after changing the tilt.

The next step was to cut the dados to match the seat tenons.  This is the same procedure as the front legs, but more complicated.  The back legs are much longer, curved, and the adder block puts a tilt into the whole business.  I found that the end of the leg was hitting the sled and I couldn’t get the adder block face flat.  To correct this, I added a scrap piece of 1/4" plywood underneath the leg.  In the instruction video, Brock has some special jigs and blocks clamped in place to hold the leg steady.  I was impatient, so I held them by hand.  This proved to be a mistake.  The first dado I cut was too small.  I should have trimmed it slowly with a chisel or plane, but instead I tried to make a little cut on the table saw.  Well, I cut too much, and that made the dado loose.  I was able to compensate by cutting the second dado very tight.  See the picture below.

Cutting the second dado on the first leg required me to move the 1/4" plywood to the bottom of the sled.  Unfortunately, I also had to raise the saw blade to account for the added height.  I cut the dados on the second leg with more care (I thought), but when I tried a dry fit with the seat, I realized that I forgot to LOWER the blade and this resulted in a dado that was too deep.  Arrrrrrgh!  I’ll now have to create a patch for this.

Lesson learned:  Take your time, and think about each step.  Measure twice.

With the dados finished (sort of), I routed the roundovers and snugged the legs into the joints.  Here are pictures of the legs, a close-up of the joint, and the chair assembled with all the legs in place.  It’s finally starting to look like a chair!

Next, I cut the front and back top of the legs at 9 degrees.  This helps the legs flow into the headrest.  Then I cut the “ears” or “horns” on each leg.  Here’s a picture of the legs after this step.

The instructions call for the arms to be cut next, but the headrest (or crest rail) looked easier, so I skipped to that step.  I started with 16/4 stock that I resawed down to about 3” thick.  I have a 14” band saw with a riser block to give me a 12” vertical cut.  I needed all of it for this resaw.  I used a 3/4" blade and it cut quite smoothly.  I jointed one edge so when I band saw the headrest curve, it will be square.  The front and back will be cut off, so no need to plane those.  The plans call for a 6 degree cut on each side of the headrest to match the angle of the back legs.  I used my compound miter saw to cut the angle.

Then the problems began…I took measurements from the plans and used a bit of trigonometry to determine the distance across the top of the headrest.  But this calculation didn’t match the headrest pattern, or the actual dimensions I was getting on the chair.  I quickly realized that the pattern was oversized in width and could be cut down.  No problem.  Why didn’t the chair dimensions match?  After some more measurements, I found the problem.  The distance from the joint to the top of the headrest on the pattern is 33”, but on the chair it measured 34”.  I had left 1” of extra material on the bottom of the back legs (for insurance), but the dados are measured from the bottom and I forgot to account for this.  So, the joint is 1” low and the result is that the headrest will now be 1” higher.  I will have to remember this when I make the spindles!  The headrest angles were now off, so those had to be cut down also.  The picture shows the headrest clamped into place.  It still needs to be customized.

Lesson learned:  If you are going to add extra material, make sure it is clearly marked.

All-in-all it wasn’t a complete disaster, but I’m not happy with my mistakes.  I hope these screw-ups help you to avoid similar problems.

My stats show that people have been viewing the blog from all over the world.  Thanks very much for that.  Please feel free to comment.  My job regularly takes me to Australia, and I see several views from there.  So, cheers to my mates down under!

Back Legs (Part 2)

I didn’t get much done today, but any progress is still progress, right?  I started by completing the rocker lamination jig.  This is made by first gluing together two pieces of plywood (I did this yesterday and let it dry overnight.)  The plywood I used is 3/4" Russian Birch.  Next, I glued on the rocker pattern and cut it out on the band saw.  My blade is getting so dull that it was more like burning through the wood than cutting.  I screwed the bottom portion to another piece of plywood to act as a backer.  The top portion was cut into several sections to use as cauls.  I cut out some of the wood on the cauls just to lighten the whole jig and to shorten up the clamping distance.  On most lamination layups I’ve seen, the clamping is done perpendicular to the curve.  My jig is setup to clamp straight up and down.  I’ll see how this works.  If I made a mistake, I can modify it later.  If anyone has experience with this, I would appreciate some advice.

I ordered a new band saw blade today.  I have a lot more cutting to do, but it will have to wait a few days until I have a decent blade to work with.  So instead of cutting the front profile on the back legs, I worked on the “adder blocks”.  The back legs on the Maloof chairs splay out at 5 degrees.  He did this with the use of special router bits to make the rear bridle joints.  The Brock design uses “adder blocks” to simplify the process without the need for special bits.  These blocks are glued to the inside of the back legs and then cut to at 6 degrees (Brock uses 6 degrees instead of 5 degrees). 

I have mixed feeling about these blocks.  They do make the build easier, but I worry that they will look like a “patch”.  I did my best to use wood that matched in color and grain direction with the legs, so maybe they will blend in.  I’ll just have to wait and see how it turns out.

I used a hand plane to clean up the area where the back legs will meet the seat.  [ I used to view hand planes as antiques used only by woodworking Neanderthals.  I mean they don’t even have motors!  But now that I’ve taken the time to learn how to setup a plane and use it properly, it’s one of my favorite tools.]  For the adder blocks, I used a piece of 8/4 stock that I had already made four square.  I cut it in half which gave me a 3 x 5 x 7/8 block.  I clamped each leg to my bench, and then glued and clamped the blocks onto the inside of each leg.  Waxed paper was placed under the legs to make sure they weren’t glued permanently to the bench.  I’ll let these dry overnight before cutting the 6 degree angles.  The use of these blocks will become clearer in future steps.