Small Projects #51: Lacewood Box with Brazilian Rosewood Tray

  • Advertise with us
Blog entry by Madmark2 posted 08-01-2021 03:36 AM 590 reads 0 times favorited 1 comment Add to Favorites Watch
« Part 50: Perfect Mitered Cube Crystal Display Case Part 51 of Small Projects series Part 52: Steampunk Unity Weather Station Done! »

SWMBO had a friend visiting us in Fla from TGWN (The Great White North) and I made this box.

SWMBO put the final finish on it.

Ready to ship.

Waiting on final finished coat of poly.

The box is from a 15/16” x 5” x 18” piece of S2S lace wood. The tray is from a 13/16” x 2” x 24” piece of Brazilian rosewood with a big worm trail about 6” from one end.

There were only two pieces of scrap from the lace wood, both from trimming the bottom to length & width. Beyond those two, small, cutoffs, all the waste from the lace wood was dust.

Lacewood box with Brazilian rosewood tray ready for finish

Continue reading for build details:

Starting with the above, thinner stock, was called for in the actual project. This is to maximize yield and to have the box and tray read “lighter”. This is essentially a zero waste project sized to fit the avaliable single piece of lumber with as close to 100% utilization as possible. As my father would say, ”You can put the scrap in your eye!”

The finished box is 6-3/8” long x 4-5/8” high x 5-1/8” wide with the tray 5-13/16” x 4-1/2” x 1-7/8”

Both pieces were resawn and then lightly planed to even out the odd 1/32”. Actual box thickness wasn’t critical as long as everything was even. The lace wood was split down the center and yielded two ~3/8” pieces. The rosewood was cut from both faces at 9/32” and with the TK blade I have a 1/16” piece of veneer left over. The rosewood was planed to .250”.

The box was built first and then the tray to fit. The tray is spaced off the bottom by four scraps of rosewood. I put the worm track to the inside on one of the spacers so not even that went to scrap.

Since the box side thickness and width isn’t specified beyond “start with even thickness and width” cutting was done “self referentially”.

I started by squaring one end of both slabs. Then I set the cross cut stop to the width of the board. I just put the random width piece against the blade and moved the miter stop over. This is the maximum width of the end.

The top, at maximum width, makes the box one side thickness wider than the lid. The ends are two sides thinner than the box, or one thickness thinner than the width of the top. Since we just set the miter stop to the max width of the board (the top), we simply use one of the pieces as a spacer next to the miter stop. This automatically subtracts the correct side thickness from the crosscut no matter what the thickness actually is! One end is cut from each slab to maximize yield.

Once the ends have been cut, the remaining stock is cut into four even lengths, two from each slab. There should be little, if any, cutoffs.

Of the four equal slabs select the best piece as the top, the worst the bottom, and the last two are the sides. Select and mark best side of all pieces.

The bottom is trimmed the same way as the ends but with both ends as self referential spacers.

The ends are then used to set the rip fence to the bottom width. The bottom has to be exactly the width of the ends.

The pieces are dry fit at this point just as a sanity check. The lid is both longer and wider than the dry fit. Tweak basics as needed.

Sliding Lid:
The trick to getting the top essentially flush with the sides, and cutting effectively, is to correctly sequence the operations, as the top and sides interact. First, rabbet both sides of and one end of the lid with a square profile half the material thickness. The rabbet should be a few thou (~0.005”) deeper than half. Not too much, or the bottom of the dado will make the sides too thin.

Without adjusting the blade depth, use the lid to set the dado spacing for the top of the sides. Set the rip fence to match the rabbet depth on the lid. It’s easy to set if you have a ZCI. Put the top side of the lid against the rip fence. Move the fence over until the edge of the bottom of the rabbet aligns with the fence side of the edge of the kerf in the ZCI.

Run two sides and one end with the bad (inside) face down.

Check to see if the lid slides in the side dados. If needed, widen the two sides first & retest. If the lid slides, DO NOT widen the end piece rabbet. The end of the lid should NOT want to go into the end dado. if there is 1/16” or more interference between the end of the lid and the dado in the end piece, widen the end dado of the interference is about 1/32”. The front lip of the lid will be sanded to fit after gluing, during final fitment.

Once the dado width is correct, raise the blade and rip thru cut the remaining end. This insures the top of the end doesn’t intrude into the dado, blocking the lid. Save the cutoff as it will be attached to the lid in a later step.

Box Assembly:
Time for another dry fit & tweak.

Once the dry fit is good, sand the inside faces to 220.

My corner assembly jig is my locked TS fence with one of those Rockler plastic “assembly corners” and a 6” Jorgensen wood clamp. The huge surface area of the block clamp locks and releases with a quarter-turn making setup and teardown a snap. Additionally both the corner and fence can be moved away from a freshly glued piece without disturbing it.

I assemble in this order:

  1. left side along side assembly block.
  2. far end (tall with dado) against fence into side
  3. bottom.
  4. near (short) end.
  5. right side

Everything dries for at least an hour before the clamps come off.

The outside is scraped of big glue clumps and sanded with 120.

While waiting for the glue to dry on the box, this is a good time to attach the push-to-open/close bar to the near end of the lid. Because of the dado, the lid is always half a side thickness too long. Subtracting 1/2 a thickness is hard to do. We know it’s going to get trimmed so we glue and clamp the offcut from the near end no more than 1/16” back from the edge. Since the offcut is being glued directly over where it was cut from, try to orient it correctly. Sand off about 1/16” from the end and the round the corners and edges of the top side for finger comfort. The bottom edges, being the glue surface, should be left square and flat.

Lid Fitment:
In a sliding lid box a “perfect” fit is where the lid slides easily and automatically locks in place when dropped in. The end dado is intentionally 1/32” to 1/16” narrower than the front lip of the lid. To get it to seat and hold, slowly fine grit sand a bevel on the lip. Test after every couple of quick passes as it’s real easy to take too much off. Once the lip can be seated (a little gap is OK as long as the lid holds in place when upended), mark and trim most of the extra off. Leave it a little proud so the final sanding will get it dead flush with the end.

Detail of lid lip fitment.

The edge of the push/pull bar may need additional sanding.

Tray Fab:
The rosewood was sliced into two sides, three bottom/divider pieces, and two ends. One of the three bottom/divider pieces is ripped 1/4” off and used as both a center divider and lift grip for removing the tray.

The better sides of the rosewood was faced up or in since the interior of the tray is more visible

Rosewood grain is beautiful.

Glue up order:

  1. left side
  2. far end
  3. both bottom pieces
  4. near end
  5. right side

The last piece was rounded on the top and eyeballed in place and tweaked until even.

Final Fab Touches:
Both pieces received a decorative 1/8”R bead around the base. This adds interest and visually raises the pieces. Although the detail on the tray is seldom seen, it serves to tie the two different species together. The use of the rosewood as spacers attached to the lacewood ties the two species together. All the rosewood was oriented with the visually most interesting side inward because the inside of the tray will be seen more often. No attempt was made at grain sequencing. On the lacewood the best side was faced out, again without sequencing.

After beading, a final 220 sanding removes the bearing marks and smooths. The edges and corners are lightly chamfered with the 5” ROS to avoid splintering.

Felt feet (Always!), shop stamp and build date complete the fab portion of the project.


Fully fitted & detailed, ready for SWMBO to finish.

Truthfully, I detest finishing. I do it because I have to. My soft finish is just wax. Hard finish is two coats of 50%-50% poly-MS and a final coat of 100% poly. The base coats are slopped on and I don’t have a lot of patience between coats.

SWMBO has previously critiqued my finishing prompting me to say “Ok, YOU do it!” She replied “Ok.” and here we are.

Finished pics soon! (I hope!)

Two coats of 50%-50% of poly-MS mix later we’ve got:

Slop coats are for penetration and coverage, not finish.

After the base coats were fully dry the outside of both the box and tray were scuffed with a 220 grit sanding sponge as prep for final finish.

SWMBO just put the final coat of full strength poly on.

Pack and ship to friend is last, last step.

-- The hump with the stump and the pump!

1 comment so far

View Rich's profile


7348 posts in 1802 days

#1 posted 08-07-2021 04:32 AM

That’s a beautiful box, Mark. Great work.

-- Half of what we read or hear about finishing is right. We just don’t know which half! — Bob Flexner

Have your say...

You must be signed in to post the comments.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics