16

Build: 1997 or 1998, Pieces: 63, Steps: 7+2
LWH: 27/24/6 studs, 20,72/18,83/4,29 cm

1 lego unit (LU) = 2mm. Colour indicates the height of the piece, here, red is 1 plate thick, green 3 (step 3) and blue 5 (step 1).

Recording technique: Up until this model, I used a Xerox to record the models, which was difficult and tiring. The reason why I didn't record by hand was that I had an idea that I had to draw all the studs to be able to discern between plates and tiles. (pieces without studs)

While doing some sketches I suddenly realized, that it didn't necessarily have to be so: Since pieces with studs are most common it was much easier to give the tiles a signature instead: The tiles were crossed signifying "no studs." 

With this change it suddenly became easier to draw than to photocopy. However, to avoid confusing tiles with bricks (and other thicknesses), I had to use colourcode, this meant that although it was now possible to build with more colours, it still wasn't possible to record these colours, and the models stayed unicoloured.

In this very simple model, everything is mounted on the 4x10 plate in the middle.

Building technique: After I'd discovered the new jumper technique, I just had to try to fix the wings in entirely new positions by both tilting and turning.

Starting with designing a new wingdesign that benefited from the jumper technique, I now had a big problem: How to tilt them? I had only tilted the wings in three models before: 04, 05 and 14, using three different techniques, that weren't optimal for this model:

I don't know precisely why I didn't choose the first one, but with the thick wing and jumpermounting, it would reach a total thickness of 8 plates resulting in poor aerodynamics, and only give an angle of 12° (instead of 26,25°). The second was very laborious (buildup), and the third one was too dangerous to clips.

Two of the steps in angling the wings: the support plate is tilted with a 1x2 hinge and the wing fixed with the 2x3 jumper technique.

I ended up tilting the wings with standard 1x2 hinges and stayed that way since there wasn't room for them to move between the central block and the 4x10 mountingplate.

Unfortunately these two anglingtechniques were way too fragile to support the weight of the wings, and the model had a tendency to fall apart when handled. The weak point was the jumpertechnique that meant that the wings were only connected to the rest of the model by as few as two studs. This however couldn't account for the fragility of the model, that seemed to fall apart only just by being looked at, and I assumed that it was caused by old and worn studs, but recently when I recreated the model in Ldraw, I found it very difficult to fit in the wings into the space between the central block and the mountingplate, which could mean that it wasn't possible to squeeze the wings in far enough originally as well, meaning that they weren't even connected properly to the few available studs.

New type of nose

Dawn of a new era: Despite all the problems this model more than hinted the enormous potential unleashed by the combination of the jumpertechnique with some sort of tilting, and prompted further development of the technique.

Despite all the problems I think this model went very well, due to the very low brickuse (0,09722 pieces per square LU) despite the high complexity (even though this naturally was connected with the inadequate structural robustness - but I think that's ok for a first try).

Aesthetically the new type of nose were an attempt to get away from the annoying trite of the diamondstyle, and fitted the wing configuration, making this model the first model in real clawstyle, with a perfect wingconfiguration where the rear part points both outwards and backwards aligning itself with the small forward wings, and the middle section points both forward and downwards - a little like a plowshare ready to cut up things that gets too close, quite menacing!