Cato

Practical notes from doing this regularly:

• Pilot and countersink everything. Screwing near the edge of a bevelled strip without a pilot hole splits solid timber far more readily than plywood.
• Stagger the fixings rather than putting them in a straight line along the grain. A row of holes on one grain line is a perforation.
• The bevel is conventionally 45 degrees, and anything from 30 to 45 works. Shallower angles pull the cabinet tighter to the wall and are less forgiving to cut accurately.
• Fix into structure, not into plasterboard. For anything heavy this is the whole game. Find the studs, and if you cannot, use fixings genuinely rated for the load rather than the ones in the drawer.
• Run the cleat long and fix into as many studs as you can reach. Spreading the load across three fixing points is worth more than any upgrade in timber.

And add a spacer strip of the same thickness at the bottom of the cabinet, so it hangs plumb instead of leaning in.

The answer that matters most on an actual shelf, though, and it is worth saying plainly: the wood is not the limiting factor.

For a domestic shelf, the load on a support is small compared with what any sensible piece of timber will carry. What fails, in order:

1. The wall fixing.
2. The screw pulling through or the timber splitting at the fixing.
3. The shelf itself sagging over its span.

Grain orientation of the support does not appear until well after all three. So spend the attention on where the support is fixed to the wall, on pilot holes, and on the shelf span — a shelf that sags in the middle is a spacing problem and the most common actual complaint.

Worth raising with the builder directly and without accusation. This is a genuinely traditional practice in some trades — people have done it for decades on fence posts and machinery sheds, and many who suggest it have never thought about where the oil goes.

"I would rather not have used oil on this given how close it is to the beach, can we price a proper exterior oil instead" usually ends the conversation in thirty seconds. It is also worth knowing what they intend to do with the oil if you decline, since somebody has to dispose of it properly either way.

The neatest answer depends on how you cut the bevel in the first place, and it is probably simpler than you fear.

If you cut it on a table saw, tip the blade to the same angle in the opposite direction, set the height so it cuts through the plywood and just skims the frame, flip the assembly and run each edge. The blade follows the bevel because you are cutting parallel to it. Two passes and it is done.

If you cut it with a circular saw and a guide, same idea — set the same angle, clamp a straight edge, and trim.

With a router, a flush trim bit follows the frame with its bearing, and on a bevelled edge that leaves the plywood proud at one face and undercut at the other. Some people accept that and finish by hand; a bevelled trim bit matching the angle does it properly.

For the repeated assembly requirement, the hardware choice matters more than the joinery.

Threaded inserts in the tops, with machine bolts through the frame, are the right answer. They can be undone and redone many times without wearing, unlike wood screws, which lose grip after a handful of cycles in the same hole — and that loosening is exactly what makes a knock-down piece feel worse each move.

Make the corner joint a bolted connection with locating dowels or a shallow rebate, so the alignment is mechanical and the bolts only clamp. Relying on bolts alone to hold alignment means it goes together slightly differently every time.

Number and mark everything on hidden faces. After the second move you will not remember which leg goes where, and getting it wrong once and forcing it is how holes get enlarged.