We began by making a simple full-size prototype with accurate angles and dimensions. We picked out an old chair that was at our dispense and were able to disguise it as the Robie 1 by attaching a long piece of cardboard with vertical dal rods to simulate the rungs in the back. This confirmed our suspicions that the 90° rigid back is uncomfortable and provides no lumbar support. Using the measurements from the dimensioned drawing, we found where the chair failed when compared to the anthropometrics of a human. With the seat at 18" high, it is 1.8" too short for the median man, 3.7" too short for the 99%, and .7" too tall for the 1%. This highlights the necessity for adjustable heights, however, we couldn't pin down how to make that work for a chair like this. Therefore, I will be drawing the chair for myself, with a seat at 20" high. The next issue deals with seat depth. At 17.9" deep, the chair is .8" too long for the median man, 2.7" too long for the 1%, and .5" too short for the 99%. However, after adding in the rule of having a fist-width between your leg and the edge of the chair, the seat is too deep for all of them. For myself, I estimate a necessary seat depth of 15". With a seat width of 15.75", it is 1.55" too wide for the median man, and 4.35" too wide for the 1%, and 1.15" too narrow for the 99%. However, the seat being wider than necessary is no problem, so I will leave it be.
The back of the chair causes a real dilemma. We needed to keep the essence of the original design while making it ergonomically sound. The wood in the back created problems until I thought of the Johnsonite Wallart line of wall base. It is sturdy, yet flexible, and comes in faux wood finishes. We can use this in combination with a horizontal lumbar