Joinery question w/ sketch-up illustration

**TDHofstetter** · 04-30-2009, 07:47 AM

That's a pretty neat design. Appeals to my eye.

The "undertop"... if it were "natural-grain solids"... wouldn't be too bad for movement issues if it were attached to the carcase in the center of each side, then allowed to move front & rear. I presume the grain would flow left to right, so all the movement would be front to rear. A single attachment point in the middle of each side would permit that. The top shouldn't be firmly attached to the legs, just allowed to slide on the cross-dado.

OR... if the sides were oriented such that the grain is VERTICAL there, and made also of natural-grain solids, then the sides would move along with the "undertop" and all would be well if the "undertop were completely attached to both the sides and the legs... PROVIDED that the glass & frame TOP were allowed to slip in either the front or rear leg dados.

The curved sides could be made from solids, provided you started with thick stock & built a swing-arm router rig to hollow it. If you did it that way, the inside of the carcase would be angular; only the outside would be curved.

OR... kerf-bent plywood would do it... OR... ordinary 1/4" ply should bend to that radius without kerf cuts... OR... the sides could be built up from veneers, over a form used for both sides to ensure that their curves would be identical.

OR... the sides COULD be built up from vertically-oriented boards, face-laminated into a butcherblock panel after bandsawing to the curve you want. I mean... lay each smallish board flat & cut the curve out of one edge, then cut another just like it, and face-laminate 'em together, then another, and so on till your stack of face-laminated boards is as tall as the sides of the nightstand are WIDE.

OR... one could even build up the sides vertically from small boards of gradually diminishing width, bandsawn at a slight angle; in this case, the sides would have horizontal grain and the "undertop" would need to be slip-mounted but the glass's frame wouldn't.

-- Tim --

Member of the
Robert "Limey" Bolton Memorial
International
Volunteer Mentorship and Assistance
Programme

**art3427** · 04-30-2009, 07:49 AM

I'm thinking a lap joint would work. That is, both the leg and the shelf would be notched so that each piece is locked in place by it's counter part. Something like space balls could be inserted into the joint to insure the panel doesn't rattle around.

art

**art3427** · 04-30-2009, 07:52 AM

I like your design. Mind if I build it?

art

**JBark** · 05-01-2009, 09:37 AM

If I were going to make this - nice looking piece by the way - I would make that solid "top" out of solid wood and join it in the front and allow it to float in the rear joint.

I'd also make the sides out of solid stock and make myself a template to reproduce the curve of the leg on to the side. I doubt I can describe this correctly...using a bearing router bit you should be able to reproduce the curve on to the side stock on the ends and then remove the material over the surface by machine and hand. This is just as problematic for movement so I would glue it at the top and allow for movement at the bottom.

Describing things online never works for me because you can't see my hands moving.

John

**Sawduster** · 05-01-2009, 11:51 AM

I'd do like John said for that "under-top". No need to notch the shelf piece for that or for the frame for the glass.

I think you could bend 1/4" thick hardwood ply into the grooves of the side pieces without needing to kerf the back unless that curve is more severe than it looks in the drawing. 1/4" ply, like its thicker cousins, is a bit less than 1/4" thick and it bends fairly easily. I'd hate to lose as much thickness from a solid piece as that would entail to cut away the convex area of the curve.

Another method that might be worth trying if you want to use solid wood for the bent side pieces would be to resaw 3 or 4 inch widths of the stock to 1/8" or 3/16" thick, and do laminated bends onto a form, then glue them up into a panel of the width needed. Could also steam bend 1/4" thick piece to glue up for the panels, but with steam bending you have to contend with more spring-back than with laminated bends.

**D Alan** · 05-01-2009, 05:01 PM

Thanks for the input everyone. As I mentioned, I'm not really intending to build this, though Art, you are more than welcome to. This is more an exercise for learning Sketch-up in tandem with understanding the mechanics and design of building wood furniture. I am looking toward a first serious project - this drawing and a couple of others I have done helps narrow the LOML's focus on what she is wanting for a pair of bed-side night stands. Once I come up with a finished illustration I'll post it for comments and suggestions as I am sure I will have LOTS of questions about what is the best way to do something... I guess this conference page would be as good as any?

Jerry, the notching of the legs for the two tops is actually intentional as a theme/design consideration, I was trying (ROTFL) to echo shapes I have seen from sources other than the U.S. and Europe, though I cannot remember the specifics of when, where, or what. Hmmm... Actually, now that I spend a moment to think about it, it bears a striking resemblance to a to a type of jewelry setting called a prong set, which makes sense since the the LOML is working on her GIA certs. Funny how the unconscious mind works.

So these solutions bring up another issue about wood movement. For the purpose of conceptual orientation, if the grain of a solid wood shape goes from left to right, and if that piece of wood is permanently affixed on the right, does this prevent movement to the right and force all movement to the left? And when that same wood is contained on two or four sides, as in the apron or face frame between the support legs, where does the movement go? What keeps the structure from being cracked by the forces of expansion? (That sounds like a bad movie plot: Cracked!! By the Forces of Expansion!)

Tim, your suggestion to: >>... build up the sides vertically from small boards of gradually diminishing width, bandsawn at a slight angle... gives me an idea for something else I was thinking of as a design element. Definitely going to file that one away.

**TDHofstetter** · 05-01-2009, 06:06 PM

When the long grain is contained & restricted from movement on one side, all the movement is towards and away from the restriction. Like... if you glue a board's edge to the side of a building like an awning, that awning will grow & shrink with humidity changes.

When a piece is constrained on all sides and the humidity increases... one of two things happen: Either one or more of the constraining objects gets pushed out of place (often destructively, breaking joints & whatever else it takes) or if the constraint is simply too great the wood fibers will crush.

That latter doesn't happen very often, though - it's incredibly hard to fully constrain wood's movement. Granite is very often "slabbed" out of a quarry by drilling holes down into the rock, driving wooden wedges into the holes, and pouring water onto the wedges. As they swell, their force of expansion breaks the granite slab off the rock wall. THAT is HARD to contain! :)

If a piece, though, is constrained on all four sides and the humidity DROPS... then one of THREE things happens. One of the constraining objects gets moved inward to follow the wood's shrinking, or if that can't happen then the constraining joint will be stressed to rupture, or else the piece of wood will split.

It's only an issue in SIDE grain, though; there's very very little movement lengthwise, along the grain. A baseball bat will grow and shrink in diameter, but its length will always be the same.

-- Tim --

Member of the
Robert "Limey" Bolton Memorial
International
Volunteer Mentorship and Assistance
Programme

**Sawduster** · 05-02-2009, 09:25 AM

Jerry, the notching of the legs for the two tops is actually intentional as a theme/design consideration, . . .

In your design one can not tell if the top is notched or the legs or both. The same effect could be had doing any of the three methods but notching only the legs to house the top and under-top would be simplest as well as providing for wood movement by gluing to only the front legs, thus forcing all movement toward the rear. If you notched the tops or did a sort of half lap with both the legs and tops being notched the notches would lock into the legs and any movement of the wood in the tops would push or pull the legs closer or further apart.

**D Alan** · 06-02-2010, 09:05 AM

I really have no plans to build this, it's just something I was doodling with. Inset glass top night stand with an Asian and Modern influence... at least I was trying for something like that. Roughly 20" high x 13" deep x 25" wide.

Now after having drawn it I am trying to figure out how it could be built. One question I have is the joinery for the table top (not the glass top framework) as it relates to wood movement. As imagined, the legs would be notched to accept the upper glass frame and the table top's edges and I assume the easiest material for the top would be a veneered plywood which should have little or no discernable movement, yeah? But what if the table top were a solid wood panel glue up? What could be done to avoid trouble with wood movement? I was thinking the glass framework could be M&T and there would be no worry about movement as long as there were about a 1/16" or so gap surrounding the glass.

Not too concerned about the face frame, more M&T with 3/4" stock. But there would be a challenge in bending the sides of the cabinet to match the curve of the legs. I would think 1/4" veneered plywood that has been grooved on the inside face to allow it to bend is one solution but what else could be done?