Wednesday, January 2, 2008

And Then I Cut It 1/8" Too Short

Thanks for coming back! We're ready to pick up the story of the corner cabinet with the events that occurred on New Year's Eve. It was a productive day, even though I didn't get everything done that I had planned to do. Still, I got enough done.

I spent the afternoon on New Year's Eve making the "ogee" molding for the corner cabinet. The image on the left is a drawing of the exact profile generated by SketchUp. As you can see, it's not really an ogee, I just don't know what to call it, and it's like an ogee, so for now, it's an ogee.

I don't know about you, but I've never seen this exact profile in any router bit catalog. In this thread on WoodNet, I asked the membership what they thought would be the best way to make this profile. Taking all of the advice together, I figured that I'd need exactly 3 router bits to make the profile: a rabbeting bit, a 1/4" radius round over bit, and a 1/4" radius cove bit. I figured I'd use the rabbeting bit to remove most of the waste & save the wear & tear on the other two bits.

Before I could get started routing the profile, though, I had to make a new router table fence. I have been using an old scrap of red oak plywood as a fence for a couple of years now. It was only 3/4" thick and about 1" wide, so it wouldn't work for this profile.

The picture at the right shows the fence I made. It's a piece of white oak plywood scrap with a 3 1/2" radius semi-circular hole cut out of it on one edge. I took two of the triangular cut-offs from the top & bottom shelves & cut them off at the same height as the fence face. These pieces both had the factory corners on them, and I verified that they were indeed square. But cutting them down & using the factory corners, I knew that the fence would end up being square to the table top.

The table is also set up for cutting the first of the rabbets.

After making the fence, I selected a piece of white oak that I had milled up the previous Friday and the poplar test piece. I ripped both of them to the same width, which enabled me to use the same exact set up for a cut on the oak workpiece after testing it on the poplar piece without having to make any adjustments.

The free version of SketchUp doesn't print images to scale, so a couple of weeks ago I took a piece of paper & drew the profile out by hand to full scale. I cut this drawing out & adhered it to the end of the poplar test piece with spray adhesive. This allowed me to gauge when a cut was set up properly.

I started by cutting two rabbets in my blank with the rabbeting bit. The picture at the right shows how the poplar board looked after I cut both of the rabbets. You'll note that I actually left more material on the board than I'd intended to. The bearings on the rabbeting bit didn't let me remove any more material, and even though I had the new fence, I didn't want to remove the bearings. I figured this was close enough.

Next, I pulled out the 1/4" radius cove bit I had bought for this purpose. I even removed the bearing on this bit to see if I could get it all the way in to the corner where the cove had to be. It was no go -- the stub that the bearing sits on wouldn't let me move the bit any closer to the proper place.

So I ran out to the ToolNut -- a local tool store, that also sells through the Internet, and bought a cove bit . The bit I bought has a 1 1/4" cutter diameter, but a 1/4" radius. I as able to get this bit right in the corner I needed it to be in.

I then used the core box bit to cut the cove. What I did was first set the bit height by lining up the end of the poplar stick & raising the big until it lined up with the arc. Then I moved the fence in place & adjusted its position until it looked like everything was lining up right. Then I clamped on the feather boards & made a partial pass. I then made some adjustments to the fence position (the height was right).

It took a couple of tries to get the bit just right. Due to the bit's diameter, I also had to reduce the router's speed. I have no idea how fast it was actually going, but I reduced the speed by one setting on the dial. I didn't get any burning & the cut was nice & smooth. The photo at above right shows the poplar board after I had finished cutting out the cove. Of course, I also ran the oak board through after finishing this cut.

Next I chucked up a 1/4" radius round over bit, with the bearing removed, to finish off the piece. It took a few tries, but I was able to get the bit's height & the fence's position set up perfectly. The photo at the left shows how the poplar test board looked after this cut was finished.

To finish the profile, I needed to change the set up & stand the board up on edge. The stub that the bearing mounts to just hit the cove when I tried setting it up while lying flat.

It took a lot of trial & error to get this set up right. The problem was I kept adjusting the fence when I should have been adjusting the bit height. Adjusting the fence had no effect at all on the cut, but I was moving it in such small increments, I didn't realize at first that the adjustments weren't working.

Eventually, I got it right. The photo at the right shows the oak work piece after I finished routing it.

The next step was to rip the profile off of this board. The poplar board came in handy again for this. I lined up the blade with the line I had drawn & glued on to on the poplar piece & set the fence. I left the profile on the cut-off side of the blade for this rip.

I had started with a piece of oak stock that was 5' long, which is about 18" longer than I need. Unfortunately, there was about 8" or so of splits & checks on one end the board and a bit of snipe on one end of the stock from a mistake I made while milling up the lumber. I wasn't worried because of all of the extra material. So I cut the bad ends off & prepared to miter.

Depending on the kind of glue you're using, the air and the wood needs to be above a minimum temperature in order to get a good glue bond. Since I planned to glue all of the pieces I had made so far, I had placed the base cabinet into the family room at the start of the day to warm up the wood. The photo at the left shows the base cabinet, lying on the cradle I made for it, in my family room with the mitered but not yet glued on base boards and the newly made ogee molding.

At this point, I proceeded to glue the base boards onto the base cabinet. I just used glue & clamps, no brads. I then waited for the glue to dry for an hour or two before doing anything more to the cabinet.

When the time was right, I roughly centered the molding on the base cabinet & marked where it met the corners. I then mitered the two ends, being careful to end up with a right end & a left end. I'm happy to report I got this part right. And the piece ended up being the right length, too.

I had to take the off-cut from the right end & cut a miter to match the other one in it. Since the molding didn't come out perfectly symmetrical, I couldn't just put the offcut from the right side on the left & vice versa, nor could I flip it round. I did manage to get the miter cut correctly.

Next I had to cut the piece for the right side to length. I put it on the cabinet in its place & got everything lined up right. I marked it's length. I brought it to the table saw. I had the miter gauge set up for a miter cut, but I needed to set it for a square cut, so I set my Osbourne EB3 back to 0°. Then I noticed that the miter gauge's aluminum fence needed to be moved, or I'd cut into it. So I put down the work piece & moved the fence.

Now I picked up the work piece, lined up my mark with the blade, and cut it to length. It was right on the line, perfect. And then I noticed that the mitered end of the work piece was lying on the table, and not in my hand next to the miter gauge fence.

My heart sank. I picked up the mitered piece & brought it to the cabinet. And sure enough, it was 1/8" too short! At this point, I called it quits for the night.

The next day, New Year's Day, I determined that I didn't have enough extra material left to remake the piece. I had enough to make the left piece. I'd have two off cuts that were both more than 5" long, but neither long enough to make another piece. So I figured that I'd just cut the left piece 1/8" too short & make it a design feature. And that's what I did.

After that, all I had left to do is to glue the ogee moldings in place. Again, I used no brads, just clamps. The photo at the right shows the base cabinet after the glue had dried & I removed the clamps.

Before anybody says anything, yes, that's glue residue in the inside upper corner under the top. Apparently I didn't get all of it off when I put the cabinet together. Because of where it is, you'll never see it unless you actually put your head into the cabinet, but I'm going to have to make an attempt to clean it up before I finish the piece.

My next task is to make the cove / crown moldings. I'm going to have to make a taller feather board of some kind to keep the workpiece firmly against the fence, since this bit is over 2" tall.

See you next time!


neil said...

Very nice Tony!!!! Love your dialog. Never thought about it but your right, we've seen plenty of 1/2round-flat-cove, but 1/2rd-flat-cove-flat-1/2round....I'm tipping my hat.

I thought it was funny when you mentioned bringing the piece into the Family room, eventually we all end up at one point in the house.

Really digg'in it Tony...Neil

Tony V said...


Thanks! I'm glad you're enjoying it. Coming from you, that means a lot.

I wish I could take credit for the design of the molding, but it's the profile that's actually on our china cabinet. Remember, I've designed this cabinet to match the china cabinet. It did make it a little harder to make the molding.

My shop is an unheated garage, and it gets too cold in there to do glue-ups properly. Believe me, I've tried. The glue turns white & chalky when I glue up in there at this time of year. And, from what I've read, that means weak joints.

I've been working on the cove / crown molding. That's going to be an intersting story.