The perfect bookshelf for your engineering textbooks…well, ok, maybe just for a couple of them :) The total load case for the design was 75 lbs, however, I have NOT verified the accuracy of the design tool, or ‘sanity check’ any of the calculated stresses. So load test at your own risk!

The brackets are intended for 3 x #6 woodscrew fasteners (or equivalent) to anchor the shelf, and the shelf sits in three cylindrical recesses on the top surface of the bracket. The brackets are designed to be printed with the top surface down.

The shelf itself is in two pieces and requires 3, half inch diameter spheres for assembly (I use 0.5 inch slingshot ammo as ball bearings). The spheres are used in the pseudo-kinematic interfaces at the midpoint of the shelves (see images). For assembly, it is easiest to assemble the ‘tension interface' first, and then use that like a ball joint to rotate the other two interfaces into location. You may find it easier to glue the sphere to one shelf, but it's not necessary. The shelves are designed to be printed with the top surface down.

Please note, the design of the shelf requires the presence of a preload to be stable (aka no load on the shelf will make it very easy for the shelf to fall apart :) ). Also, the more evenly distributed the load on the shelf, the better. Isolated loads SHOULD be fine, but distributed is preferable.

I printed all parts in PETG. A stiffer material might be preferable, as there is some deflection at the support ends that a stiffer material probably wouldn't have (although with more load across the full surface this issue went away also). I used a 50% infill NOTE: the GD solver assumes full density, isotropic materials. I did increase the Safety Factor to account for this, but infill type and quality of layer adhesion will certainly impact this in ways that may exceed these assumptions.