wk6: computer controlled machining

Ergo Chair

For this week's assignment, I decided to be practical and make something I could use in normal everyday life. I was inspired by the many strange ergonomic chairs out there, and chose to make my own wobbly ergonomic stool.

Design

The primary design constraint for my chair was that it needed to have a round bottom for maximum wobble. The chair would also need to be big enough for me to sit on and use at a desk. Ideally, it would be made using press-fit construction.

Creating enough wobble was one of the biggest challenges. I first considered only making two curved rails for the chair legs—a glorified rocking chair. But I really wanted to maximize the degrees of freedom for the wobble. I ended up using four rails: two in the x-direction and two in the y-direction. This configuration allowed for maximum stability at the neutral (vertical) postion as well as stable rocking motion in the x and y directions.

It took me three tries to model this design reasonably well in Fusion. I used several parameters, construction lines, and components to generate this model. This turned out to be a good choice because I had to change my material thickness from 0.43" to 0.5" parametrically before setting up my cut file.

I made the mistake of creating each type of chair leg as one component (i.e. two legs in the x-component, and two legs in the y-component). This turned out to be a problem when laying out my cut file. I ended up exporting a dxf of my fusion geometry to Rhino and setting up my cut file from there. I added a cap with alignment holes for the top of my stool to cover the unsightly dogbone joints.

Mill

After setting up my file, Calvin helped me generate the G-Code from MasterCam. We used the 3/8" compression endmill to cut the material in one pass with a 0.005" onion skin. We ran the job on the ONSRUD in N51. Unfortunately, the onion skin was insufficient on this job. Not only did the endmill cut all the way through the OSD, but several of my pieces lost suction with the vaccuum bed and had to be thrown away.

Calvin helped me set up the file again in MasterCam. This time we set the onion skin to 0.01" and only used half the bed with full suction. This job ran successfully.

Post Process

The OSD required A LOT of post processing. I spent several hours with the electric sander sanding down rough edges and resurfacing the OSD. My notches turned out to be slightly too snug for press fit joints so I had to file them down and hammer the pieces together. In my design file I accounted for 0.5" material thickness and 0.005" tolerance along all edges of a joint. I probably should have made the tolerance a little bigger because the material thickness varied from 0.49" to 0.53" when I measured.

The most challenging part of assembly was putting the top of the stool on the legs. It was really difficult to get all eight leg posts into the correct holes. I think one of my legs must have been cut incorrectly when the bed lost suction in the first job because it was almost 0.5" out of alignment with the top piece. I sanded, filed and hammered these pieces until they stuck together. Hopefully the stool will not implode from internal strain.

Finally, I glued the cap piece (no dogbone joints) on top of the stool. I aligned it with the layer below using bolts through the alignment holes. I placed a sacrificial piece of material on top of that (no glue) so that the clamps would not dent the final surface. I let the glue set overnight, removed the clamps and did a final sand of the seat and edges.

I'm pretty happy with the result overall. The stool is less wobbily than a real (expensive) ergo chair. But suprisingly, it is possible to rock in any direction on this stool. Now I just need to make a cushion or something for more comfort.