mahogany headboard

[Rob]

The current project is a mahogany and leopardwood headboard for our king sized bed. I'm trying to finalize some dimensions right now so I can proceed with making sawdust. So far I've only trimmed my rails to rough length. I need to figure the final height out, and then everything will fall into place from there I think.

Right now I just have pictures of pretty sticks. The one on the right is leopardwood, this pic doesn't show the dramatic grain at all. You'll see it soon enough....

[Rob]

Okay, work commenced today in earnest on this project. I'm taking a break to let my bandsaw motor cool down before making a rip cut, so I'll slap up some pics and a walkthrough of work done so far. 

Last night, I built two sawhorses at a 37" height. These were made from cull bin lumber from Home Depot that I jointed and thicknessed so it's all purty, but still cheap. Each sawhorse has about $3 in materials. the beams are douglas fir 2x12s, nice and beefy. The height was set so that I could use a cheap, slightly damaged lauan door I bought at Lowe's for 5 bucks as a quick&dirty outfeed table for my table saw. The reason why will be apparent shortly. here's my lowbuck outfeed table:


Okay, the first order of business with the outfeed table done was to rip the rough edge off my 8/4 mahogany stock for posts. I bought this S3S, or surfaced on 3 sides. So I found the narrowest point and set my rip width for about a whisker and a half less than that. Here's the rough surface I'm removing:


...and here's the most expensive 2x4 I ever bought  , with the rough face ripped off:

My saw handled it well with minimal burning. I did make a dumb mistake though, I forgot to attach the shop vac to the dust collection port. So even though I turned on the vac, I got a fantastic orange dust cloud. 

Next I did a similar rip to cut off the rough edge of my leopardwood stock. Since I promised you some grain pics, here they are. In one I wiped the board down with a little naphtha to better show how it will look with finish applied. Note the check in the end of the board. I am planning to cut that out, but I can't get rid of all of it. 




That was a success, so I was encouraged to try what would possibly be the hardest operation in the project next.  I am using the leopard wood to make panels. The leopard wood is 3/4" thick and 4" wide, so to wind up with panels 1/4" thick and 8" wide, I'm resawing, which means using a bandsaw to slice the board down the middle of its thickness. My bandsaw is 1/2HP, so it's not ideal for this operation. Luckily, I have a good blade, and have been practicing my resaw (all those boxes!) technique. So I set up for the cut as shown in the pics below. I'm using a 1/2" 3TPI blade. thicker would be better, but my 12" saw can't tension a thicker blade properly.




With the setup done and featherboard tightened to hold hte board against the fence, I ran a scrap of cherry I had to test the setup. Wow! it worked and looked great, and it will wind up being used in a future project, I'll guarantee you that. This is what's cool about bookmatching--you get a symmetrical grain pattern since you slice right down the center of a board.



Both halves looked equal thickness within very acceptable tolerance, so I was thrilled. Time to slice the expensive (and really dense) stuff. I crosscut the 6 foot length to three equal boards each the length of two panels plus one saw kerf width. This was just to make them easier to handle on the bandsaw table.


Then the moment of truth. More like 15 minutes of truth--I did a superb job on the first one. The second one wandered off track and jammed, but I freed the blade by wedging a thin scrap in the kerf. Then I flipped it and started from the other end, meeting the original cut in the middle somewhere. The third one went okay, but not as slick as the first one. All will give me usable 1/4" panels though, which is a relief. I had no margin for error here--no extra stock to play with and the lumberyard was out of anything that would even come close to matching (I went yesterday). If I messed up this resaw, I would have to redesign my panels. But it all worked out.

Okay, then I crosscut my post stock to rough length while my motor cools off. I put a plywood scrap behind it to keep the end grain from tearing out when I crosscut it.


The post stock is one of the straightest, clearest boards I've ever seen of any type. I have about two feet of offcut, it will make some incredible boxes.

Next operation is to rip the post stock right down the center to get two posts the same length. When they are thicknessed they will be 1-5/8" square. The width of the stock is 3-3/4, so I have a little bit of room for error, but not a lot. So this rip will be done on the bandsaw. I have the fence and featherboard all set up, I took dozens of measurements with a caliper and I'm within a few thousandths of dead center of the board.

And that brings you up to date. I'm going to chug a cuppa coffee and hop back down to the shop.

[Rob]

Some people stopped by and I had to stop, but I did get time to rip my post stock into two (which worked fantastically well on the bandsaw) and plane the saw marks off everything I've ripped and resawed so far. Here's a couple shots of the post stock immediately after ripping. The sawn face is up on both sticks.



After a couple planer passes, you get it looking like this. This nicely shows what  mean by removing saw marks.

and finally after all planing:


I was worried sick about tearout on the leopard wood. Its grain is so non-straight that I figured a planer would tear the hell out of it. Happily, my resawn pieces were substantially thicker than I needed, so even though I didn't have any scrap left to practice on, I got to dial in my planer technique for this stuff.

I've now removed all the saw marks from my panel stock, but I haven't done the final thicknessing passes in the planer. I plan to switch to new knives first and generally clean up the machine before I try to finish. My elevation wheel raises or lowers the cutter 1/16" for a full turn. By taking passes of 1/12 of a turn (right around 0.005"), I was able to plane these things to a glassy smooth surface that is almost entirely free of any tearout. 

The fresh surface is a lot brighter orange than the exposed wood I removed. And the grain is much more prominent, and the surface is glassy (this wood is really hard).




I'm short on time (I have to return to work Tuesday, and we're traveling Monday), so I aim to get the following done tonight before I close up shop:
change out my planer knives
do the final thicknessing of the panel stock to .250 (or a little under)
build my glue up jig/cauls/fixture/whatever
get glue on the panels so I can cut them to finished size tomorrow.

The panel stock is currently 24+ inches by 4" wide. I'll glue two 4" widths together to get 8", then cut that down to two 8x12-inch panels. Total will be six panels, so three glueups. These will fit into a slot in the rails so that the final apparent dimension will be 7-1/2 x 11-1/2". 7.5x12 is real close to golden mean, but I didn't have quite enough length in my leopardwood thanks to that check in the end.

[Rob]

I aim to get the following done tonight before I close up shop:
change out my planer knives check
do the final thicknessing of the panel stock to .250 (or a little under) check
build my glue up jig/cauls/fixture/whatever check
get glue on the panels so I can cut them to finished size tomorrow. check

Here's a quick shot of a glueup jig with the panels in place and a caul on top to put a gentle flattening pressure on the panel. The base is a 2x2 foot scrap of melamine from the cull bin at HD. Strips of 1/2" ply spaced to the exact width of the panel are nailed to the base. Then the panel halves sort of "snap" into place when you push them down between the strips.

[Rob]

Meet boardzilla.



I (my wife) decided that the posts made from 8/4 were too insubsantial. My options were to laminate, or trek back to the yard for some 16/4 stock (12/4 isn't available at my yard for some reason, and 16/4 isn't available in any other species but basswood). So I went the 16/4 route.

I returned with this monstrosity: 14 board feet, weighing 46 pounds.



Even my shop cleaning assistant doesn't weigh that much:


The board coast about as much per pound as lean ground beef, so I guess it was a good deal.

Boards that big are hard to handle and hard to cut. Luckily I had considered this beforehand. My strategy was to crosscut the stock to rough length, then rip it into two square posts on the bandsaw. The crosscut was actually the more challenging cut. Luckily, my saw has sliding rails and a sliding crosscut table. Sliding the rails all the way left and using the sliding table, I could push this board right at its balance point, which is a pretty well controlled cut. But for a board this heavy I added support at the end of the extended rails by propping them up on a sawhorse, so the whole shebang didn't tip under the weight (my saw is not cast iron). I took the crosscut in five passes, and at the full height of my blade I just broke the top surface of the 3-1/2" thick board.  By clamping on a scrap of plywood, I avoided tearout on the exit.





The bandsaw rip was not especially interesting or challenging, except that I bought a new 3tpi blade for insurance. It barely slowed down in the 3-1/2" straight grained mahogany, so it was mainly an exercise in keeping the 40-pound stick square to the fence and table. My wife asssisted, keeping hte edge against the fence for me at the beginning of the cut while I fed the board from behind the balance point. Resawing the 4" leopard was a lot harder than this rip was.

Then I took the ripped post that had one rough edge to the jointer, and squared it off using one of the faces against the fence for reference. Moving a 20-pound stick is still strenuous, but actually my jointer confidence built quickly as I realized the mass of the thing sort of worked in my favor. Its own weight kept it on the bed pretty well.

Now I used the thickness planer to get to a state where I had four faces planed to my liking. I got down to about a 3-inch thickness on one by removing some worm holes I didn't like. So I'm not there yet, but finished post size is going to be in the 3 inch ballpark, which is a substantial-looking post. The 8/4 will get used when I make nightstands to match.



Later tonight I'll:
+ take these to final thickness
+ lay out the mortises for the rails
+ surface the rail stock, which I've already cut to rough length
+ determine the final width of my panel field, based on six 7-1/2" panels and set the stile width from that (it'll be in the 5 inch ballpark)

If I can get all that done tonight, I'll be happy. The remaining steps would be:
+ rip and crosscut the stiles
+ rip rails to finished width
+ setup the router table for the grooves in the rails and sides of stiles.
+ route the above grooves
+ mortise the posts
+ route the tenons on the rails and stiles
+ dry fit (woohoo)
+ glue it up and clamp like mad

About a full day's work if all goes according to plan, and I have two days to get er done, plus I just found out we have a paid day off Tuesday since the market is closed, so I'm now not worried about completing this. The router work is actually more comfortable for me than the saw operations were, so I feel like it's a downhill trip to the finish now.

I got a new router bit: at the yard they had a 1/4" solid carbide dual flute spiral cutter on the clearance rack. It was pretty much tailor made for slotting the rails. It's not long, but it only needs to make a 1/4" deep slot.

[Rob]

I did finish squaring up the posts and planing them to final dimensions. I also selected top. bottom, a forward-facing side, and a side to take the mortises. I marked the pieces accordingly. Here you can see my checking them witha try square. It took lotsa passes on the jointer to get back to square after I removed so much with the planer earlier.



In my final few passes through the planer, I paid a lot of attention to squareness. Here's a shot of each post while I measure the thickness of the face with the mortises:



Close enough! Actually, checking at several points up and down the length, they are thicknessed accurately to within 0.005 (10 times less precise than the pics would suggest), so I'm happy.

One thing I left off the to do list is to clean up the panels and crosscut them to their finished height. To recap, I started with 1 stick that was 3/4x4x73", which I then:
a) crosscut twice into three, two-foot + sticks
b) resawed each stick into 3/8 thick or so stock
c) thicknessed all six to 1/4"
d) edge jointed and glued up into three 8x24+ panels
e) cleaned up the glue line with a card scraper
f) crosscut to six 8x12 panels
So from 3/4x4x73" to 6 @ 1/4x8x12 in six easy steps (not so easy in fact).


And lastly, while I was browsing tonight I saw something that gave me an idea for what to do with the offcut (over 2 board feet!) from boardzilla:

But that will be an upcoming major project of its own!

[Rob]

Last night's progess: I didn't get a trremendous amount of work done, but I did hammer out several design items that required some calculation and thought.

I face jointed and glued a couple cutoffs from the rail stock to get a thicker board. I'm going to use this to make a cap for each post. This echoes a design element (the stair railing posts) in our hallway. To make it match even more closely, I'll make a cove molding like the posts for the rail have.

To register these baords for the glueup, I used a trick that I'm getting real fond of. I took one of my 2x2foot melamine scraps, and used my brad nailer to tack some MDF strips on. I arranged the strips to register to three sides of my glueup pieces. A little packing tape on the MDF ensures a good release, otherwise the glue squeezeout would attach them to the work. The first pic shows my prepared glueup jig. The second shows how I clamped it in my clamping sawhorse/workbench gizmo. I don't have a pic of the glue being applied (because you don't want to lose much time) but instead of a roller like you always see used, I put the glue on with a plastic bondo squeegee, which worked superbly to give me a thin, even layer over the whole surface.



Here's a picture of the panels laid out between rail stock in approximately the final spacing. Instead of space, there will be mahogany stiles between the panels, but this gives you an idea of the look of the panel field at least.


Then I set up my lowbuck outfeed table and ripped the rough edges off the rail stock. Nothing special there, but heres a pic of the ripping setup. You know how everyone says "guard removed for clarity" on their pics? I'll be honest and say "guard removed because I don't use it." Narrowest rip hereis still over 5 inches wide, and while I'm not complacent about safety, that's a pretty reasonable cut. They all ripped with only the slightest of burning.

The rest of the night I spent finalizing design. There are a very few critical dimensions in this piece, and my work so far has kept all the options open possible, but at this point I have to commit to some dimensions to proceed, so it's time to do that. Here's how the design is developed.

Things I can't change:
king mattress width: 76"
matress top height: 28"
max post height: 61"
post width: 3"
panel field height: 11-1/2" (the panels are 12" but will be housed in a 1/4" deep slot inthe rails)

Almost everythng else is pretty well up for grabs. I made two design decisions with input from the bed's other intended occupant. First, the posts should be spaced so that the centerlines are at 76". This keeps open the option of making a footboard and actual bed rails and converting this to a complete bed later if we choose. Keeping options open is good. Second, we decided that some rail needed to extend below the mattress height, so that pillows don't fall behind the headboard, and the edge of the board isn't ever touching your head. Below that, the space will be open between the posts, except that a stretcher rail will be placed a foot or so from the bottom to keep things all square and rigid.

The first decision establishes a required finished length for my rail stock: 76 inches minus two halves of a post width (3"), plus the length of the tenons. I'll set the tenon length at 1 inch for the moment, wich I think is conservative. So my finished length to crosscut the rail stock is 75". 1" mortise depth is a working assumption, and I can make them shallow if need be.

Now the stile width needs to be set. I'd like it to be in the 5 inch ballpark, since I have a 5-inch wide board set aside for it.  I'll need two half-stiles, which should be close to half the width of a stile, but can take up some of the slop if it's convenient to do so (so I don't get stile width like 4-57/64 or whatever). If the stile width is S, and my half-stile width is H, I start out with
73 = 6*7.5 + 5*S + 2*H
6 * 7.5 is the panel width, 45". So removing that from both sides,
28 = 5*S + 2*H
If I set S at 5, then
28 = 25 + 2*H so H = 1.5. A true half-stile would be 2.5, so these are narrow by an inch. that's a little narrower than I want really. Let's try a quarter inch narrower stile:
28 = 5*4.75 + 2*H so 2*H  = 4.25 and H = 2.125 (2-1/8). Half of 4.75 is 2-3/8 so we're just a quarter inch shy here.
Just for fun, let's set S at 4.5.
28 = 5*4.5 + 2*H so 2*H = 5.5 and H = 2.75. this would make them a little too wide I think.
Split the difference and set S = 4.625 (4-5/8), and you get pretty close: H = 2.4375 (2-7/16)

I hate working with measurements, so I'm going to go with S=4.75. I need to make sure my half-stile have a tenon on on end too, so I'll have to rip them down to final width with a separate setup. stiles will need to be 12" in length to fit in the slots in the rails. So the final dimensions are:
stiles: 4-3/4 x 12"
half-stiles: 2-3/8 x 12 (but add the tenon on so rough dimension is 3-3/8 x 12")
panels: 8 x 12" but we already knew that. They're made.


I need to set the depth of the panel recess relative to the face of the rails. I can slot down the center of the rail, this will be the max recess, about 1/4" (rails are 3/4 thick and panels 1/4, give or take a few thousandths). Slotting closer to the front face will allow me to increase the apparent thickness of the panels. I don't want to slot closer than 1/8" to the face though.

The thing about not centering the grooves is that I'll have to make two setups to route the tenons on the stiles, since the face-to-tenon cheek distances will differ. Centering the slot give me one setup (with the added bonus that my error is always doubled  thanks to symmetry). I won't really ever need to repeat the setup for tenoning the stiles though. Once I get the front cheek cut, I can "dial in" on the back side bit height, and tweak with sandpaper if I need to. And there' no real need for the measurement to be critical--I can dial in whatever recess looks right to me with some rail scrap. I think something in the 1/8" area is appropriate, perhaps 3/16".

Now I need to set the height of the panel field and each rail. Let's assume that I will start from the 26" height and build upward. I have the widths of three rails (imaginatively designiated R1, R2, and R3 from top to bottom) and the effective height of the panels and stiles, call it P. My top of my stack will fall at 26 + R3 + R2 + P + R1.

Right now, I have R1 = 6-1/2, R2 = 5-1/2, R3 = 5-1/2, and P= 11-1/2 as noted above.
26 + 5.5 + 5.5 + 11.5 + 6.5 = 55", which is well shy of my 61" post height. I'm probably aiming for the top edge of the top rail at 59", so that my post caps have a little height above the rail, but 59 - 55 leaves me 4 inches shy of the goal.

The design tradeoff is: I can buy another board to fill up the gap, or I can drop the top down by those 4 inches and use the stock on hand. I could also try a little of each, or I could allow some space between rails 2 and 3.

I think the look I want is best accomplished with buying more stock; adding rail R4 and keeping the top edge at 59" will give me the most visual impact. why do I need so much height? I have this in a room with a cathedral ceiling for one, and the missus like to stack pillows up, which made me raise the panel field up to the height where it is in the first place. She can have her pillows, and I can have my leopard wood, and all are happy. But I'm a board short. I''ll have to pick that up at the yard on Tuesday.

I won't have much shop time today or tomorrow. I have other things to get done today, adn tomorrow will be a travel day, but I can pick up the work on Tuesday, and work evenings from there on to finish this. this is getting real close, I know it looks like I've just been turning boards into different sized boards (and that's pretty much what I have been doing). But all that dimensioning will make this go together pretty easily when it's time.

The updated to do list then:
+ lay out the mortises for the rails
+ surface the rail stock
+ rip and crosscut the stiles
+ setup the router table for the grooves in the rails and sides of stiles.
+ route the above grooves
+ mortise the posts
+ route the tenons on the rails and stiles (I have several options here, and this may be where I spend some time fine-tuning a setup)
+ make molding for the cap pieces
+ cut caps to final size

Stuff that will have to wait till I have the R4 in hand:
+ dry fit (woohoo)
+ glue it up and clamp like mad

[Rob]

Quick rollup of what got done:

Prior to today I was mainly working on my jigs and fixtures--I built a new router table and a tenoning fixture. This morning was the first I made some mahogany sawdust in a while. I futzed around as long as I could (I installed more shop lights, swept the shop, moved stuff around, and generally avoided the project).

Today I got out a clean notebook and planned the steps necessary to complete this thing. And then started tackling those steps.
+ I thicknessed all the 4/4 stock to .750, +/- 0.010 (I was shooting for .005, and mostly got it)
+ I finished cutting every piece to finished dimensions
+ I mortised the posts (scary!)
+ I tenoned the ends of all the rails (really scary, but worked as slick as you please)

I've got pics, but I'll save them for tomorrow, I'm sleeeepy--I had to correct above where I typed tenononon..." 

[Grump]

This is gonna be really pretty. That's some great lookin' wood.

[Rob]

Even the sawdust is purty.

[Rob]

Not much to describe here, I did this with my plunge router (Hitachi M12V, 3-1/4 HP) using a 1/4" spiral upcut bit (solid carbide). One thing was that I reference the back face fo the posts, so I didn't have to be exact in centering the mortise--any error would cancel. This generated absurd amounts of dust.















Wound up about .255 wide and a little more than an inch deep. My tenons will be an inch, so the extra depth can catch excess glue. The width is onthe high side due to bit cahtter. I really wish I could have found a 1/2" shank bit, but this had to suffice. My 1/4" straight cutter wouldn't give me the full 1" depth I needed.

[Rob]

Again, this isn't a huge discussion, just I used the fixture I built to cut these tenons. In a couple cases where the ends were a few thousandths short, I built them back up with packing tape--since the bearing rides the board face, I can add back about 3 thousandths at a time of thickness this way toget my tenons closer to the target.








test piece fit was a little loose (but not much), so I made adjustments


nice looking tenon!



the hard part was remembering to climb cut when using this fixture. Normally you wouldn't, but this jig makes it impossible to screw up too badly. By climb cutting, you avoid the end grain tearout. I get some fairly serious tearout on the pieces where I cut in the ordinary direction. For other reasons, I'm remaking the top rail, but it was the worst for tearout in addition to being th most visible. I eventually marked a big arrow on the tabletop to remind me.

[Rob]

Next I had to slot the rails to accept the panels and stiles. Here's some pics of the router table setup:





I added a featherboard to keep the rail tight to the fence, it's not shown here. I messed up the top rail in the process of figuring out that this was necessary. 

Here's a couple more pics from just dry-fitting a couple parts together to check fit.

[Rob]

Finally I can show you something that almost looks like the finished project. this is an "almost" dry fit--I have to fine tune some tenons or figure out a better assembly technique or something. The stiles aren't fitted into the slot in the bottom rail here. They're still a wee bit snug.

Add a couple 3" posts at the outside of this and you have the finished project, more or less.



The half stiles on the outside haven't been tenoned to fit into the post mortises yet. I may do some fine-tuning there before I commit to a final width. i have the cumulative error over six feet to factor in. I'm happy that at this point the panel field is a little wider than the rails--if it were narrower I'd have more work to do.

[Grump]

It's a shame to put something so beautiful in a room you spend most of your time with your eyes closed.