Very much like the way we constructed the profiles, we have tried to devise the best way to set them up to ensure that the overall geometry is accurate from the beginning. Our first step was to create a beam suspended at a height similar to that of the profiles.
That beam is then notched at 14" apart to receive the profiles so that we can stand them upright. The notched beam acts as a jig to ensure that the profiles are held in the precise position to receive the interconnecting structural elements. We also repainted portions of the wood platform to show the plan positioning of the profiles.
Once in place, we began the laborious tasks of measuring, cutting, welding, and grinding of the steel pieces that connect the profiles together. One person will then mark the position of the steel piece with blue tape, while another person positions the piece correctly and cuts the ends to the exact angle. They are then welded together.
We also have one person in charge of double-checking to see if all of the pieces have been made, constantly moving back and forth between the digital model and the real thing. We learned that the sheer number of elements has made it relatively easy to miss a few. Knowing that once the frames are painted it becomes much more difficult to make corrections, we are making every attempt to find the errors now rather than later. So far we have averaged one error per four-bay module (which is not bad…or is it?).
Simultaneous to the making of the wall pieces, we have also started to build the horizontal seating element. This piece is definitely the most difficult element geometrically, because we are not just connecting single pieces of steel section between two profiles. In fact, the bench is composed of multiple, three-dimensional sections of steel frames that extend out of one of the bays of the wall. We have had to map the points of intersection of the entire bench both in plan and in three-dimensional space in order to build it correctly.
I think we are feeling pretty good about how the steel fabrication is going, however, we are really starting to get concerned about the roping process. Next week we have decided that we will try to completely finish and paint one of the four-bay modules so that we can experiment with the roping process early on rather than waiting to finish all of the steel work before starting.