The story continues

Thanks to a good working session with my tutors I finally have a modeling approach that gives me the control I want. Here is a first experiment with it finding out how the shadow might look, made using 'Project' in Grasshopper. (Sun movement is reverse from natural btw.)
In this model I can set radius, height & width of the cell wall and angle of the corners in the bottom row of the structure. It then determines the nr. of cells around and in height on its own to make a fitting roof.

Movement controll

The movement of the cells can be controlled in two ways, the first is to change the angle of the corners. The twist in the cell walls stays constant.

Turris movement by cell twisting

The second way the movement of the cells can be controlled is by changing the twist in the cell walls the angle between the cell walls stays constant.

Intermediate animation result

After some further work and finally making a tower out of the modeled cell this is the result, an impression of the final roof structure. However there is still a fault in the propagation of the cells for they now cross eachother a little bit. You can see this in the corner point where little square overlaps form. However it serves as a good illustration of my goal.

cell modeling done

 After too much time I managed to model the cell using vectors as a base.

The method is both the most perfect and the easiest one I've tried, after realizing how vectors work. Thanks Andrew for now calling me stupid for I must have appeared that way. 

Too bad I spent a week on the others that failed. 

But now I can finally continue in modeling the full TurrisMotus, but first some sleep.