With the legs firmly seated and set in each specific mount, we had ourselves a tall and goofy stool. All that is missing a the seat back to make this thing a chair. The legs are arranged such that their is a slight tipping backward so when one sits down they are actually tipped back into the seat back, into the repose that the chair itself invites.
I knew that the back was going to be the most difficult part of this build. It is a single curved piece made up of a couple different laminations. The original chair had roughly a 3/16" thick back. It was remarkable how well it's held up which made it seem all the more well thought out and well tested (and making our efforts that much more critical). A big theme surrounding this chair was that if they could make it 60 years ago that there should be no reason why we can't remake it today with all our fancy gadgetry. Turns out that this is a great way to ramp up the pressure and stress we applied to ourselves.
Several attempts at making a mold for creating the back were tried and failed miserably. I was desperate to make just two successful backs to get these prototypes done but this was unsettling because one-offs were fine for one-of-a-king pieces but disastrous for attempting to test an actual production method.
I turned back to the CNC crew. They sculpted a perfect three dimensional negative from the model and produced a plug that was perfect in everyday. But there was a huge problem with our approach. The seat back is not only curved but it also sits into a curved and angled groove. What is brilliant about the original back is that the thin lamination they used also remained fairly flexible. This aided their efforts in manipulating the back into the groove which also helped to create the best feature of the chair, its sculptural shape where the back takes on a conical feature as it deflects. Our challenge, to increase the thickness of the back with another layer of 1/8" core material. To double this back thickness and still be able to put it into the groove worried me.
We made a new mold. We eliminated the third axis and essentially cut semi circles that stacked up in a length. The third axis had to happen in the groove and so we gambled on a shape for the mold that was sort of halfway between the top of the back where the curve was mellow and the bottom of the back where the curve was severe.
It worked. Beyond the actual veneer, we actually pulled off the experiment. A replica chair made completely different with a thicker seat back and custom aluminum leg attachments.