Milling the big board

Well, we're finally starting to make things that roughly resemble bed parts.

Yesterday, +Jill Van Patten was sick, so I was in the shop by myself for the day.  I worked on some more templates, then started shaping the footboard legs.  I start by taking the milled lumber for the legs, then I screw the templates to that material.  I carefully locate the screw location so that the screw holes will be covered by one of the 100 or so ebony plugs that will be added once the footboard is glued up and sanded.  I then trace the outline of the template with a permanent marker, remove the template, and rough cut the material on the bandsaw.  Then I screw the template back on and use a flush trim bit in the router table to make the leg perfectly mirror the template.

The joint where the legs meet the top rail on the footboard and headboard is a splined miter joint.  The first step is to cut a perfect 45 degree miter on both the legs and the top rail.  This must be done before the legs or top rail are shaped, so that we have nice flat and square faces to register the 45 degree angle.  At this point, the top rail is oversized, which is fine for now.  Right now, we're just getting the angle correct.

Here's a shot of how the legs and top rail of the footboard will join together.

Here's a closer picture of the joint between the top rail (which is on the left) and the left leg (which is on the bottom).  You can see some of the shaping detail on the outside portion of the leg (the very bottom in this picture).  This detail is an example of what took so much time in getting the templates right.

In the above picture, that flare-out at the joint is not permanent.  I put that there because it's the only way to clamp the rail and the leg together during glue-up; once the joint is glued, I'll cut that off.  You'll see that better when the time comes in a future post (hopefully not too far into the future).

Today, I tackled one of the most physically and logistically challenging tasks on the bed.  The bottom rail on the footboard is a ginormous piece of material.  The finished board is 10-1/8" wide, 1-1/2" thick, and 67-5/8" long.  I only have one board in my lumber storage that is big enough for this beast, and it's a monster.  13" wide, 2-1/4" thick, and nearly 10 feet long.  Keep in mind that mahogany is much heavier than pine or Douglas fir.  Milling and jointery on something this large tests the limits of my tools and the physical limits of my arms.  Milling this monster board flat, square, and to final dimensions took me nearly five hours.  I start by rough-cutting the material oversized.  I begin at the miter saw.

I have a 12" sliding compound miter saw, and it is just barely has the cutting depth to handle this board.  Note the stand supporting the left side of the board, and the hold-down that normally goes unused got some exercise holding this board down during the cut.  Just setting everything up for this simple cut took about 10 minutes.  Maneuvering the material is a huge hassle.

Next, I trim the board to a slightly oversized width at the bandsaw.  Remember that the material here is 2-1/4" thick.  When I brought the board over to the bandsaw, my little 1/8" bandsaw blade ran and hid.  So I installed a 1" resaw blade which was better suited to the task.  I had to set up roller stands on both the infeed and outfeed side of the bandsaw, because there's no way I could hold this board while making the cuts.  

That 18" bandsaw is a large saw, weighing over 500 lbs.  But the board makes it look a bit undersized.  It also makes my garage look undersized -- I had to open the door to have room to feed the board through the saw.

Just as I started feeding the board through the saw, I noticed a staple in the end of the board, right where the blade would have cut  That's a quick way to ruin a $150 bandsaw blade.  There's no good reason to put metal staples in the end of the boards, and it frustrates me when lumber retailers do that.  And by the time they paint it all blue, it's hard to see the staples.  Plastic staples would be perfectly suitable, safe, and do less damage to blades.

So the board presented significant challenges for my poor bandsaw.  But ultimately, as between the board and the saw, the saw won:

So now we have the board rough-cut to slightly oversized dimensions.

Now it's time to start getting it flat, square, and to finished dimensions.  First step is the jointer.  My fairly large 12" jointer just barely has the capacity to face joint an 11" wide board.  After a few passes, you can see some areas starting to flatten and other areas where we still have sawmill marks:

And subsequent passes lead to a more finished surface:

 The final shot shows just a tiny bit of sawmill marks, but those can be removed at the planer.  The surface is flat enough to ensure that there is a reference surface to ensure perfect results at the planer.

Now we see the final surface of a very large board:

About halfway through the milling process, the sawdust from this board had filled up the 15 gallon drum of my dust collector.

By the time I finished at the planer, the 15-gallon drum had filled up a second time.  More than half of that enormous board is now sawdust, and by the time I'm done with it, the finished board will be about 1/3 the weight of the original monster.

My planer has a spiral cutterhead, which leaves a nearly flawless finish.  But to make sure the final surface is perfect, I save the last .015" for my wide belt sander.  This is one of the best tools ever invented.  I run through a couple passes on each side with 100 grit sandpaper, and in about 2 minutes I save myself hours of sanding down the road.  The pic disappeared from my camera, but I'll take another for a future post.

The last milling steps were to joint one edge of the board on the jointer and then cut it to width at the tablesaw.  Unfortunately, the board had a large enough crown that it was taking a long time to edge-joint at the jointer.  And this board is still heavy enough that maneuvering it on the jointer, picking it up, carrying back to the infeed table on the jointer, and repeating, was giving me quite a workout.  So I clamped it in my workbench and trimmed it with a handplane.

After a little work with the handplane, I only had 3 more passes on the jointer before the edge was perfectly flat and square.  One run on the tablesaw, and it was finished.

This process took roughly 5 hours.  Each cut goes fairly quickly, but for each cut, it takes quite a bit of time to set up the equipment, the support stands, and all the other things you have to do to make your tool handle this large of a board.

Enough milling for one day.  On to more interesting things, like joinery.  Classic woodworking wisdom says to do your joinery in the morning, because you get a bit tired and careless in the afternoon.  So today I pushed my luck, and fortunately didn't do any damage.  The legs and bottom rail will connect by a very large mortise and tenon joint.  I begin by cutting the tenons using my multi-router.   The multi-router is a machine that allows precise x/y/z movement against a router bit -- in this case, a 1/2" bit that I'm using to cut a 1/2" x 6" mortise in the footboard legs.

I normally use the multi-router to cut tenons as well.  I began cutting a tenon on the bottom rail.  However, it turns out the rail is way too big for my multi-router, and I couldn't finish the tenon in the usual way.  So I closed up shop and I'm taking the evening to contemplate other ways to cut the tenon.  There are many options, but they each have their own complications when you're dealing with material as large as the rail.  I'll probably go with some combination of the multi-router and my tablesaw, but we'll see.

Jill spent the afternoon making ebony plugs.  The plugs are small, decorative ebony cubes that are pillowed (rounded) on the top and chamfered on the bottom, and each one will be inlayed into a square mortise in the bed.  There are over 125 ebony plugs on the bed.  They each must be sanded, cut, and polished by hand.

Tomorrow, more joinery. And then lots and lots of shaping.

Milling and template making

+Jill Van Patten and I spent some more time in the shop, right up until Christmas Eve dinner.

First, we continued milling (and milling and milling) lumber.  Here are a couple gratuitous shots of me at the planer:

Now, I know what you're thinking:  Where can someone get a sexy leather apron like that?  Well, seriously, you can't just expect me to throw information like that out there.  If word gets out, then pretty soon everyone would be wearing one, and next thing you know it wouldn't be cool anymore.  It would be that whole star-bellied sneetch thing all over again.

We've now milled up all the lumber for the legs and top rails, and we'll save the other parts for later.  And not just because I have a short attention span.  In the past, I've tried milling up all the lumber for a project at once.  Bad things happen.  Sooner or later, you get confused about which lumber was supposed to be for which part, and then suddenly you don't have the right lumber for all the parts and you have to go back and mill up a bunch of lumber again.  So it's best to build a few parts at a time.  Plus, there's the whole ADD issue.

And so we moved on to template making.  Jill was not thrilled about making templates.  I can understand why.  You spend a lot of time in the shop and never seem to make progress on building the actual bed.  But building good templates are critical when you have pieces that have all these weird shapes.  There's no way to build multiple pieces that match if you don't have a template to follow.  The templates begin with drawing parts on poster board (or on several poster boards taped together because they don't make poster board in pieces as big as a king size bed).  The drawings have to be perfect, because the template will never be any better than the drawings.  Then, you cut out the drawing with an xacto knife and trace your poster-board template onto some 3/4" MDF. It's not rocket science, but it does have to be perfect.

Once the template is traced onto the MDF, we cut it out with a bandsaw.  Now comes the monotonous part.  The bandsaw does not leave an accurate or fine edge, so we have to turn those rough edges into perfectly straight edges.  We do that by taking a straight piece of MDF and attaching it to our template with the straight edge where our finished line should go, using double-stick tape.  Then we use a flush-trim bit in our router table to get a perfectly straight line on the template.

This takes a long time because there are many short, straight edges on these templates that are connected by weird curves and things.  And we have to route each straight edge one at a time.

Once the straight edges are done, it's on to the curves.  This is where the old-fashioned hand tools come in:  rasps and files, taking a smidgen at a time until the curve is perfect.  Here are some of the final templates (or mostly final, anyway):

Next step, cutting real wood!

Starting the Gamble-Style Bed Project

Jill and I are spending our Christmas vacation building a bedframe.  After 22 years of marriage, Jill has this crazy idea that we've outgrown the metal bedframe that comes with the mattress, and now it's time for us to have a grown-up bed.  Now, some people might think that sounds totally kooky.  But being the perceptive husband that I am, I recognize that Jill is not really asking for a bedframe.  What she really wants is for me to start a cool woodworking project.  What a lucky man I am, having a wife who is always on the lookout for ways to keep me entertained.  So we set aside our Christmas break to build a bed.  This will be the first time Jill has worked on a major project with me in the woodshop.

I decided to document this project because I thought it might be interesting to some of my friends who may not know much about furniture making.  There's more to it than many people realize, so hopefully this will entertain you, and maybe show you a few things you didn't know.

This project is inspired by a bed designed by Greene & Greene for the Gamble house in Pasedena:

Greene & Greene were two brothers who became fairly famous architects in Southern California.  They blended the arts & crafts style with Japanese woodworking tradition.  You don't see many commercially made pieces in the Greene & Greene style because the details and joinery are somewhat complex and time consuming to build, but for that same reason, many woodworkers are drawn to it.  

One of the first challenges in this project is translating the design from a twin bed to a king.  The original Greene & Greene design was perfectly proportioned, and if we simply widen everything, it will look ridiculous.  In fact, I once saw some photos of a queen bed someone had built by stretching this twin design until it was wide enough for a queen mattress, and it lost all sense of balance and proportion.  It was like when you watch a full-screen movie, but you stretch it out to the width of your widescreen TV, and all of the characters are short and fat and their cars all look like some giant person stepped on them.  We can't have that.  On the other hand, we also can't turn the twin bed into a king simply by increasing all of the dimensions (horizontal and vertical), or we'll have a footboard that stands 6 feet high, a headboard that goes through the ceiling, and a mattress so high you need a ladder to get into bed.  The challenge is to widen the bed, keep it about the same height, but adjust the pieces and details in a way that maintains the overall balance and proportion.  Sadly, I'm not that talented.  But happily, I know people who are.  Which is just as good, really.  And with the help of their artistic flair, voila, I've now got a design for a king size bed that would make the Greene brothers proud.  (I'm not going to post the drawings; you'll just have to keep reading this blog and await the finished product if you want to see how it looks in the end.)

And so, design in hand, we start milling lumber.  We're building this bed out of African Mahogany, which we buy in large, rough-sawn planks.  This one is about 13" wide by 10 feet long by 2" thick.  We started with many boards like this (some wider).

The first step is to turn big boards into smaller (but still pretty large) boards.  These will be the legs for the headboard and footboard:

Here are the top rails for the headboard and footboard:

"Rough-sawn" lumber is lumber that comes straight off the huge bandsaw they use at the sawmill to slice trees into boards.  It shows bandsaw marks and is nothing close to flat.  There are lots of splinters on it.  For a project like this bed, it will take dozens of hours to mill the rough-sawn lumber into flat, square, pretty boards that are dimensioned and ready for joinery.  You can buy pre-milled lumber, but by the time the lumber moves from the sawmill to the lumberyard, sits at the lumberyard for weeks, then moves to your shop, it goes through so many humidity changes that it will be warped and twisted.  We need the lumber to be perfectly flat and square.  So we mill it ourselves.  Also, this allows Jill to brag about how she spent her Christmas vacation muscling large planks of mahogany over our 12" jointer.  Which makes her pretty much the coolest mom in town.

Speaking of the 12" jointer, this project is testing the limits of our tools.  That doesn't happen often.  That jointer is 12" wide and over seven feet long.  It's a much bigger jointer than any sane person has in their garage.  In the photo above, Jill is jointing a board that is 3' long, 4-1/2" wide, and 2" thick, and the jointer makes that board look like a little toy.  But some of the larger boards for this project are a real challenge, even on a jointer this large.  We had to do some woodworking gymnastics to make the huge boards work on these fairly substantial tools.  Beds are big.  King sized beds are really big.  As another example, here's a picture of my clamp rack:

It's difficult to see in the picture, but this rack is double-sided -- there's a wall of clamps on the side facing us, and another wall of clamps on the opposite side.  But with all those clamps, I don't have a single clamp that will be long enough to clamp up this bed when we're ready to glue it up.  Darn.  I guess that means I'll have to buy some more clamps.  Which is the purpose of having woodworking tools -- so you can take on projects that give you an excuse to buy more woodworking tools.

Back to the jointer.  The real purpose of the jointer is to make one face of the board perfectly flat, and one edge of the board perfectly flat and perfectly square to the  face.  When the face of a board is perfectly flat, we say that the face is "true."  When the edge of a board is square to the face, we say that it is "tried."  When the face is flat and the edge is flat and square, we say that the board is "tried and true."  And that's the etymology of the phrase "tried and true."  See, I told you that you would learn something reading this blog.

Once we are finished at the jointer, we move to the planer.  The purpose of the planer is to make the unjointed face of the board perfectly flat and parallel to the jointed face, and to thickness the board.  We finished planing the top rails for the headboard and footboard just before we closed up the shop for dinner.  

In this next photo, look carefully at the grain lines going along the length of the two boards.  See how the grain curves up and down, as if the grain has kind of a hump in the middle of each board?   It's difficult to see while the wood is still raw, but it is definitely visible, and it will be much more pronounced once we apply a finish.

One of the artistic aspects of woodworking is finding boards, and sections of boards, that have grain and color that will complement the structure of the finished piece.  If you look in the top photo at the original Gamble house bed, you'll see that the top rails on the footboard and headboard have a graceful arc, rising up to a slight crest in the middle.  That arc is a critical part of the design -- go back and look at that picture again, and imagine how different the bed would look if those top rails were just straight boards with no arc.  When you have a design element like that arc that is so essential, you want the grain of the wood to accentuate the arc.These boards do that, because the arc of the grain generally follows the arc of the top rail. Looking again at the two boards in the photo immediately above, also notice how the grain is very similar from one board to the other.  This is because they were cut from the same board (so they were right next to each other in the tree).  As a result, the grain in the top rail of the headboard will match the grain in the top rail of the footboard.  Most people will never consciously notice that the grain follows the arc, or that the footboard grain matches the headboard grain.  But at a subconscious level, your brain picks up all of this.  And these details are what separate a piece of art from a construction project.

So, as we were picking out boards for the top rails, we looked for boards that had this type of grain pattern.  In a perfect world, we would have found grain that curved more gradually, but you don't get to tell the wood what grain to have.  It can be difficult to find a board that is at least 12" wide (so that you have enough material to get two top rails), seven feet long, and 2" thick, plus have exactly the grain pattern you are looking for.  I consider us very lucky to have found this type of arc in a board large enough to make our rails. In fact, in the original Greene & Greene Gamble house bed, the grain in the headboard top rail and footboard top rail match almost perfectly, and they were both clearly cut from the same board; however, the grain does not arch, but rather slants upward and to the right throughout the boards.  So the guys who built that bed (the Hall brothers, who are fairly famous woodworkers from the early 1900s) were thinking the same thing I was thinking, but they weren't lucky enough to find a board with such a cool arcing grain pattern.

Enough bragging about the cool boards we found.  The other part of the project we started today was creating approximately one boatload of jigs and templates.  Take a careful look at that first picture of the Gamble bed, and notice that there are very few 90 degree angles.  There are only eight perfectly square boards on the entire bed (two of which are the side rails).  Everything else has angles, curves, multiple curves, cloud lifts, and all sorts of weird things.  Each of which requires a jig or a template, or both.  We'll probably spend as much time making templates and jigs as we spend building the bed itself.

The first step of template making is to make a full-scale drawing on poster board (more precisely, on several pieces of posterboard taped together, because most of the pieces on this bed are very large).  There are lots of weird curves and angles in this bed, so just drawing it up takes a lot of time.  I spent about three hours making full-scale drawings of the legs and top rails.  Unfortunately, the pencil lines don't show up at all in my photos, but you can see three poster boards taped together to give me something long enough to draw some of the large pieces we're dealing with.

Tomorrow, I'll pick up some MDF and transfer the poster-board drawings to the MDF for templates.  More pictures to follow.

It's going to take a while before any of this starts looking like a bed.  The typical woodworking experience is that about 50 percent of the time is spent milling and cutting joinery, then 10% is spent making cool details and gluing it all up (which is when it starts looking like the thing it's supposed to be), and then the other 40% of the time is spent sanding and applying the finish, and wishing you were done already so you can move onto the next project.