This week I designed a work tray for using my computer in bed. If you search on google, you can find something like what I'm describing, but none of these products is exactly what I'm looking for. Most notably, non of them have a cupholder that is secure enough for near-computer use.
Originally I planned on doing some test cuts with the OSB material, so I created a set of profiles in rhino to test out the dimensions and design (I ended up bailing on this idea and went to the lumber store instead). Since the OSB falls apart if it is cut too thin, I kept the thickness of the tray at least 1.5" throughout. This is definitely more bulky than I'd like the final product to be.
Some useful commands for this design process were:
Intersect - generate point at intersection of two curves
DeleteSubCrv - remove part of a selected curve between two points
Fillet - create fillet between to curves - used this to remove sharp corners from profile
Mirror- copy and mirror objects across a plane
Join- join curves into closed loops (fixes any gaps that might be there)
PlanarSrf - create planar surface from a closed loop of curves
ExtrudeSrf - extrude a surface into a 3D shape
BooleanDifference - subtract one surface or polysurface from another to create cutouts
I added half inch rad fillets to all corners to remove any sharp geometry from the piece.
I measured a bunch of the glasses in my house and found that the bottom diameter of all the cups was at least 2.5" and the top diameter was at least 3", going up to 3.5". A nalgene is about 3.5" in diameter. I kept all this in mind when generating the dimensions of the cupholder.
I used a 0.01" offset for the moving piece of the tray to allow for movement.
1.5" cuts are moderately bendy
2.5" cuts are very bendy
0.5" cuts are more rigid
This piece has a few curved surfaces on it, and I'd like to try out a living hinge that follows this curvature. The amount of curvature allowed by a living hinge is dependent on a few factors - the width and depth of cuts, the density, the direction of the grain, type of wood, etc.
I designed a set of flexure tests in illustrator to be cut on either the laser cutter or the desktop shopbot (since the other shopbot was busy this week). For the laser tests, I drew 1/4" and 1/8" notches to simulate the types of cuts a shopbot would make (I never actually used these files since the Beam and Trotec lasers were having exhaust issues). In the shopbot tests I drew a series of straight lines of various lengths with 1/8" spacing in between; I cut these out with an 1/8" endmill on the Desktop Shopbot.
I found that patterns with longer cuts allowed for more curving of the wood - this makes sense because it has the effect of removing more material from the board. I tested this on three different lengths of cuts - 2.5", 1.5", and 0.5", the amount of allowed curvature is shown in the images above.
I also learned about some of the complications of cutting flexures on the shopbot. One thing I did not anticipate is how difficult it is to securely fixture the material to the shopbot as it becomes more and more flexible. The image below shows how sloppy some of my cuts became towards the end, when the material was flexing a lot each time the end mill cut into it.
I paused my job often to apply more tape and try to secure the material better.
After I picked out some nicer 1/2" and 1/8" baltic birch ply and learned something about the limits of flexure joints, I re-desinged my model in Solidworks with my new material in mind. This time I kept a min feature size of 1" for all my profiles so that the 0.5" ply would stay stable/flat. The design uses finger joints and slots to assemble the main structure of the tray. The 1/8" flexure cut ply is glued on top of the frame and inside the small shelf.