04.06.09 - Structure Received...

Cycled over to Leicester Laser to navigate their delivery van back to DMU. Accepted delivery of 36 unique plywood elements, transported within their negative frames. Elements stored flatpack in spaces left in the sheets they were cut from. Shock as it is discovered that the decision has been taken to move the structure from its design location (outside on a wall) to a new location (inside, suspended from a ceiling). Whilst this makes sense in that we no longer have to treat the plywood or worry about vandalism, it causes problems in that - structurally speaking - it was not designed to 'hang'. There would be problems with turning due to the imbalance around the centre of gravity. Had this decision been made before fabrication, we could easily have adjusted the design to 'hang'. This is an advantage of the parametric design solution. The joint system, weighting and curvature could all have been adjusted with simple 'sliders' in Grasshopper and the CAD drawings - that serve as instruction to the laser-cutter - automatically updated using the same software.
Consultation with DMU's technicians, Estates department and workshop staff has to wait until Monday morning. Design team concerned with the 'look' of the new joint whilst other concerns - such as the ever-increasing proximity of deadline - may take necessary priority.

03.06.09 - Prototype Mk2 / Fabrication

Painting the room continued today.

The second 1:5 prototype was finished this morning, and despite some slight production problems, it fitted together just fine. This means that I can go to the cutters' later happy that the structure will fit together.

The modelshop uses a laser-cutter that is controlled via Corel and we encountered problems importing the nets as blocks from AutoCAD 2007. These included a random and arbitrary scaling of certain pieces to an unknown and slightly different scale from that set, and the misplacing, mis-alligning or corruption of text produced using the 'annotate' function in AutoCAD. Ray was able to to fix this manually, but it is less than ideal.

The 4mm ply also has a tendency to warp after cutting as it takes on moisture. Robin - the woodshop manager - assures me that this shouldn't be a a problem in the 12mm version as it exhibits grater stability.

The structure is now in four distinct sections - each of which can be mounted individually and is self-supporting. This should aid the tricky operation of attaching the final full-scale structure to the outside wall.

Before setting off for the fabricator's to test full-size pieces, the model was assembled in the studio. We can use the model to test the insertion of various models / drawings / photographs into its 'bays' - and also to serve as an assembly diagram during full-scale production.

At the fabricators (Leicester Laser) we ran some test pieces to check finish quality. The plywood sheets were flat enough to be clamped into position in the cutter and the cut of the laser itself produced a satisfactory finish. There were a few issues with the quality of the 'etch'...but these were sorted out quickly. We were able to increase the distance between the laser and the cutting bed, which in turn increased the width of the cutting beam producing an etch visible from a distance of 4 or 5 metres. Were the plywood shows grain on its face, the laser cuts through pith and resin with differing extent. This resulted in an uneven edge to the etchings. We solved this by increasing the depth to which the laser 'etched' - cutting a channel rather than merely scorching the surface. We were also able to test the tolerance on the slotted joints, which turned out to be pretty tight - ideal for our purposes.

With a satisfactory test obtained the job was left with the fabricators until tomorrow.

02.06.09 - Amending the Design...

Painting the walls started.

Final nets were outputted in order to be tested in the morning - we will build a second prototype before fabrication now we know that actual production time will take less than a day at the fabricators. Working closely with the fabricators has meant that we are in a position to make decisions like this.

01.06.09 - Evaluating Prototype Mk1

With the defective piece re-cut we were able to assess the first prototype.

We found that the distribution of joints was unstable in some places. It was decided to re-organise the system of joints so that the structure is made up of four independent sections - each stable in its own right. We were still restricted in the length of horizontal rib we could achieve. This is defined by the clear diagonal distance across an 8ft x 4ft sheet of plywood, including the 35mm and 10mm borders needed for clamping during fabrication.

The nets were adjusted by hand due to time constraints. We would like to have been able to work out the automatic update script and we will do for future versions - it is definitely possible.

29.05.09 - Prototype Mk1

With the script finally finished we sent the first set of nets to our in-house laser-cutter to make a prototype from 4mm plywood. Our cutter can take 600mmx300mm sheets. Using Corel to read nets from AutoCAD has resulted in one piece being scaled by an arbitrary and random amount - this means we are unable to test the prototype until we get the piece re-cut on Monday. We could however test aspects of the design such as the text alignment and joint tolerances. We had tried aligning straight edges with the edges of the plywood sheet in order to reduce the amount of cutting (and therefore production time) but we discovered that the sheet must be clamped into position. This is likely to be necessary at full scale also.

28.05.09

27.05.09