Research Process
︎︎︎ back
Documentation of the research process that lead to the study “Workflow of designing surfaces with distinct kinematic properties”. The creation of parametric models supports an understanding of natural patterns, while the classification of natural systems helps to distinguish between constructed patterns.
︎︎︎ back
Documentation of the research process that lead to the study “Workflow of designing surfaces with distinct kinematic properties”. The creation of parametric models supports an understanding of natural patterns, while the classification of natural systems helps to distinguish between constructed patterns.
Grasshopper and Kangaroo exploration
1.
Closest Circle Packeging on a 3D surface
2.
Guided Circle Packeging on 2D surface (radius depends on Image colour)
3.
Textile Simulation with tile physics (trial and error)
1.
Closest Circle Packeging on a 3D surface
2.
Guided Circle Packeging on 2D surface (radius depends on Image colour)
3.
Textile Simulation with tile physics (trial and error)
Pattern deformation is relevant to the curvature the 3D print creates
The animation demonstrates how the Tessellation results from the deformation of the base-grid. A regular set of points is arranged in a triangular grid. One point in the grid can be chosen to push the others gradually to the side. The amount of push and the radius can be controlled. When the position change of each point in the grid is tracked a force vector can be illustrated. The area where the force vectors are more dense most transformative energy is stored. While the area of the tile that covers the most amount of points locks the transformative energy, as for a textile print the plastic blocks the stretching back of the textile. So the areas with the highest density of force vectors will have the most potential for the textile to stretch back.
The second animation shows how a PLA print on textile behaves. the area where less textile is covered will stretch back most intensely, resulting in one guided 3D surface deformation.
The animation demonstrates how the Tessellation results from the deformation of the base-grid. A regular set of points is arranged in a triangular grid. One point in the grid can be chosen to push the others gradually to the side. The amount of push and the radius can be controlled. When the position change of each point in the grid is tracked a force vector can be illustrated. The area where the force vectors are more dense most transformative energy is stored. While the area of the tile that covers the most amount of points locks the transformative energy, as for a textile print the plastic blocks the stretching back of the textile. So the areas with the highest density of force vectors will have the most potential for the textile to stretch back.
The second animation shows how a PLA print on textile behaves. the area where less textile is covered will stretch back most intensely, resulting in one guided 3D surface deformation.
21-07-20
![]()
![]()
Evomimetics “What are we optimizing for?”
Discussion around D. Adriaens, "Evomimetics: the biomimetic design thinking 2.0"
The focus of this paper is on the pitfalls of the general assumption that nature optimizes through evolution, by showing that this assumption is prone to generate false ideals towards optimization of Bio-Inspired design. Adrianes states that engineers misunderstand evolution for a process of optimization, while evolutionary processes are seen as a complex balance from many different parameters that are all together forced (through natural selection) to develop into one functional organism. Natural organisms therefore present „just good enough“ or „better is not possible“ solutions within a given framework of complex parameters (ecosystem etc.)
Adrianes sees great potential through collaborations between engineers and evolutionary biologists. „Already understanding how individual species solved biological problems can be instrumental for researchers in BID, whereas, working with evolutionary biologists to study entire evolutionary lineages of species can provide a more generic understanding of how form and function dynamically relates to a changing environment.“Adrianes claims that Evomemetics is a workflow where engineers together with evolutionary biologists could extract overall functional principles that can later artificially (with the tools of the engineers) be optimized to perform maximum. The paper looks at optimization as the pure mechanical performance of a design.
From a Design perspective, this approach misses some key conceptual elements. As Designers get more and more involved with the cycle of production and circularity, products are commonly understood as a complex balance (and compromise) between a variety of parameters (we also use the word „ecosystem“ here). The concept of one advanced optimized parameter (for example lightweight construction modeling) is a classic goal of engineering disciplines. Since the 70’s parallely to the introduction of the concept „Wicked Problems“ by Rittel & Webber Design started to look more generally at complex problems. This approach could be made productive as one additional component in the suggested Evomemetic workflow described by Adrianes. Could the question „What are we optimizing for?“ help to advance the workflow? For instance skyscrapers proof that men made structures can expand natural properties (due to the engineering paradigm of optimizing a small parameter space). Even though skyscrapers are a „just good enough“ solution for densifying the economic value of property in the city center, as it would make no sense to build a skyscraper in the village. This example should illustrate how needs and requirements towards certain materials are often following old economic mechanisms. The approach of looking more broadly on optimization and even more intensely on the processes in nature that allow for balancing out a huge variety of needs could be a suggestion to arrive at new solutions and developments.
Discussion around D. Adriaens, "Evomimetics: the biomimetic design thinking 2.0"
The focus of this paper is on the pitfalls of the general assumption that nature optimizes through evolution, by showing that this assumption is prone to generate false ideals towards optimization of Bio-Inspired design. Adrianes states that engineers misunderstand evolution for a process of optimization, while evolutionary processes are seen as a complex balance from many different parameters that are all together forced (through natural selection) to develop into one functional organism. Natural organisms therefore present „just good enough“ or „better is not possible“ solutions within a given framework of complex parameters (ecosystem etc.)
Adrianes sees great potential through collaborations between engineers and evolutionary biologists. „Already understanding how individual species solved biological problems can be instrumental for researchers in BID, whereas, working with evolutionary biologists to study entire evolutionary lineages of species can provide a more generic understanding of how form and function dynamically relates to a changing environment.“Adrianes claims that Evomemetics is a workflow where engineers together with evolutionary biologists could extract overall functional principles that can later artificially (with the tools of the engineers) be optimized to perform maximum. The paper looks at optimization as the pure mechanical performance of a design.
From a Design perspective, this approach misses some key conceptual elements. As Designers get more and more involved with the cycle of production and circularity, products are commonly understood as a complex balance (and compromise) between a variety of parameters (we also use the word „ecosystem“ here). The concept of one advanced optimized parameter (for example lightweight construction modeling) is a classic goal of engineering disciplines. Since the 70’s parallely to the introduction of the concept „Wicked Problems“ by Rittel & Webber Design started to look more generally at complex problems. This approach could be made productive as one additional component in the suggested Evomemetic workflow described by Adrianes. Could the question „What are we optimizing for?“ help to advance the workflow? For instance skyscrapers proof that men made structures can expand natural properties (due to the engineering paradigm of optimizing a small parameter space). Even though skyscrapers are a „just good enough“ solution for densifying the economic value of property in the city center, as it would make no sense to build a skyscraper in the village. This example should illustrate how needs and requirements towards certain materials are often following old economic mechanisms. The approach of looking more broadly on optimization and even more intensely on the processes in nature that allow for balancing out a huge variety of needs could be a suggestion to arrive at new solutions and developments.
23-06-20
![]()
Topology Optimization (Part 1), 2018 by Tony Abbey
Topology Optimization (Part 1), 2018 by Tony Abbey
Modeling and Simulation Method
During our regular group meetings, we commonly developed a method towards modeling and simulating biological Tessellation Systems. The circular graphic on the right shows a four-step method where each ring illustrates a step of the process. The steps are thought to be evolved in an iterative loop starting from the insight out and then jumping back in the center.
1. A biological specimen is selected due to comparative analysis of its evolutionary and biological context.
2. the specimen is digitalized with a CT scanning workflow, the digital model is cleaned and abstracted with software such as Amira.
3. The digital model is analyzed and systematically abstracted (rebuild in in Rhino / Grasshopper environment). This step should reveal a parametric space of the biological specimen.
4. The digital model is virtually simulated and parametrically iterated to relate form to function, most interesting results will be translated to physical prototypes exploring the physical properties of the generated geometries.
During our regular group meetings, we commonly developed a method towards modeling and simulating biological Tessellation Systems. The circular graphic on the right shows a four-step method where each ring illustrates a step of the process. The steps are thought to be evolved in an iterative loop starting from the insight out and then jumping back in the center.
1. A biological specimen is selected due to comparative analysis of its evolutionary and biological context.
2. the specimen is digitalized with a CT scanning workflow, the digital model is cleaned and abstracted with software such as Amira.
3. The digital model is analyzed and systematically abstracted (rebuild in in Rhino / Grasshopper environment). This step should reveal a parametric space of the biological specimen.
4. The digital model is virtually simulated and parametrically iterated to relate form to function, most interesting results will be translated to physical prototypes exploring the physical properties of the generated geometries.
11-06-20
![]()
Bistabale Systems through pre-stretching
This is a series of experiments investigating the ability of textile as interface material to control the 3D kinematics of a 2D tessellation pattern. The pattern is created in Grasshopper, a regular triangular grid of points is used as a starting pattern. The points are used as Centers for a Voronoi tessellation, resulting in a regular hexagonal grid. The grid of points is then manipulated from the center of the system, pushing the points gradually away from the center. In a second step, the offset of each Voronoi cell is defined by its density factor (cells that are more tightly packed get a higher offset than cells that are more loosely packed). I will explain the parameter space and function to derive to this pattern in a separate post.
In this post, I want to concentrate on the ability to control one and the same pattern (formation of hard Tessare) due to tension trapped inside a textile surface (soft interface of the system). The textile surface is used to fulfill two functions at the same time: 1st it simulates the soft interface between the hard Tessare (simular to tessellated material systems we find in nature) 2nd it traps the tension that later applies to the whole system. All samples are created with the same workflow. The textile I used is described in the post below as Stretch Net I, a 15x15 cm membrane is pre-stretched and clamped to the building plate of the 3D printer (Zortrax M200), a TPU material (85 shore) is used to simulate the harder Tessare in the Tessellated System. The TPU is extruded directly on the textile surface (after pretension), the layer hight of the print is 3mm. I will now explain the differences between pretension and the resulting function, ranging from multi-stability to bi-stability.
This is a series of experiments investigating the ability of textile as interface material to control the 3D kinematics of a 2D tessellation pattern. The pattern is created in Grasshopper, a regular triangular grid of points is used as a starting pattern. The points are used as Centers for a Voronoi tessellation, resulting in a regular hexagonal grid. The grid of points is then manipulated from the center of the system, pushing the points gradually away from the center. In a second step, the offset of each Voronoi cell is defined by its density factor (cells that are more tightly packed get a higher offset than cells that are more loosely packed). I will explain the parameter space and function to derive to this pattern in a separate post.
In this post, I want to concentrate on the ability to control one and the same pattern (formation of hard Tessare) due to tension trapped inside a textile surface (soft interface of the system). The textile surface is used to fulfill two functions at the same time: 1st it simulates the soft interface between the hard Tessare (simular to tessellated material systems we find in nature) 2nd it traps the tension that later applies to the whole system. All samples are created with the same workflow. The textile I used is described in the post below as Stretch Net I, a 15x15 cm membrane is pre-stretched and clamped to the building plate of the 3D printer (Zortrax M200), a TPU material (85 shore) is used to simulate the harder Tessare in the Tessellated System. The TPU is extruded directly on the textile surface (after pretension), the layer hight of the print is 3mm. I will now explain the differences between pretension and the resulting function, ranging from multi-stability to bi-stability.
11-05-20
Sample a.) presents a bi-axial prestretch in the membrane, the textile is stretched from its original sample size 15x15 cm to 25x25 cm, the tessellated pattern covers an area of 8x8 cm. Once the membrane is released the textile wants to reorganize and stretches back to its original size of 15x15 cm. While the larger Tessare allows less of a stretch back the smaller Tessare move closer together, resulting in a sphere-like expression of the surface system.
Sample b.) presents a uni-axial prestretch, the sample is stretched from the original size to 15x25 cm. Once the textile is released the tension in the textile forces the whole system to react differently to sample a.). An oval cylinder-like expression pops out of the surface.
The production of sample c.) differs from the samples a. and b. as two layers of pre-stretched membranes are introduced in one system. One membrane is pre-stretched exactly as described in b. while the second membrane is pre-stretched in the other axes (25x15cm). This sample shows a really interesting behavior as it appears to be bi-stable, with two energy lows (symmetric energy curve) but still able to express two different shapes. Once the surface is pushed in one direction the system reacts to the tensions created by the first textile, resulting in a horizontal rolling. But when the system is pushed in the opposite direction, the tension of the second textile starts to control the deformation, expressing a vertical rolling.
Sample b.) presents a uni-axial prestretch, the sample is stretched from the original size to 15x25 cm. Once the textile is released the tension in the textile forces the whole system to react differently to sample a.). An oval cylinder-like expression pops out of the surface.
The production of sample c.) differs from the samples a. and b. as two layers of pre-stretched membranes are introduced in one system. One membrane is pre-stretched exactly as described in b. while the second membrane is pre-stretched in the other axes (25x15cm). This sample shows a really interesting behavior as it appears to be bi-stable, with two energy lows (symmetric energy curve) but still able to express two different shapes. Once the surface is pushed in one direction the system reacts to the tensions created by the first textile, resulting in a horizontal rolling. But when the system is pushed in the opposite direction, the tension of the second textile starts to control the deformation, expressing a vertical rolling.
a![]()
b
![]()
c![]()
b
c
a
![]()
b
![]()
c![]()
b
c
Material Samples
Stretch Net I - bi elastic, 70g/qm, 89% Nylon, 11% Elasthan
Stretch Net II - bi elastic, 60g/qm, 82% Nylon, 18% Elasthan
Power Net - bi elastic, 134g/qm, 85,4% Nylon, 14,5% Elasthan
Insect Net - uni elastic, 45g/qm, 100% Polyester
Net robust - non elastic, 106g/qm, 100% Polyester
Stretch Net I - bi elastic, 70g/qm, 89% Nylon, 11% Elasthan
Stretch Net II - bi elastic, 60g/qm, 82% Nylon, 18% Elasthan
Power Net - bi elastic, 134g/qm, 85,4% Nylon, 14,5% Elasthan
Insect Net - uni elastic, 45g/qm, 100% Polyester
Net robust - non elastic, 106g/qm, 100% Polyester
04-05-20
![]()
Simular Projects
List of related and similar research projects investigating shape-changing materials, active and prescribed kinematic states and inflatable structures.
List of related and similar research projects investigating shape-changing materials, active and prescribed kinematic states and inflatable structures.
II
![]()
Self Forming Structures
[Gabriel Fields - Nervous Systems]
[Gabriel Fields - Nervous Systems]
SpaceSymmetryStructure
[Daniel Piker]
[Daniel Piker]
Numerical simulation of the pattern formation of the springtail cuticle nanostructures
[A.E. Filippov, A. Kovalev, S.N. Gorb]
[A.E. Filippov, A. Kovalev, S.N. Gorb]
Review Presentation
The presentation is organized in four parts: 1st introducing the geometrical principles and rules described by Peter Pearce, 2nd how the rules relate to the construction of patterns with Grasshopper, 3rd approaches that translate the virtual patterns into physical prototypes with material boundaries and 4th similar approaches found in related design and engineer based investigations and research projects.
See the Keynote presentation here
The presentation is organized in four parts: 1st introducing the geometrical principles and rules described by Peter Pearce, 2nd how the rules relate to the construction of patterns with Grasshopper, 3rd approaches that translate the virtual patterns into physical prototypes with material boundaries and 4th similar approaches found in related design and engineer based investigations and research projects.
See the Keynote presentation here
16-04-20
![]()
Double Membrane & 3D Tessare
This is an experiment exploring shape change through 3D tiling. The third dimension (and tile angle) pre-defines the amount in which the system can expand and shrink. The 3D tiles are interconnected on top and button via two pre-stretched membranes. Pressuring the top part can influence the whole system and the surface shapes three-dimensionally (negative Gaussian curvature) when pushing the button the state flips into positive Gaussian curvature.
This is an experiment exploring shape change through 3D tiling. The third dimension (and tile angle) pre-defines the amount in which the system can expand and shrink. The 3D tiles are interconnected on top and button via two pre-stretched membranes. Pressuring the top part can influence the whole system and the surface shapes three-dimensionally (negative Gaussian curvature) when pushing the button the state flips into positive Gaussian curvature.
04-04-20
Active inflatable Tessellation Systems (morphables)
How does the surface curve when the cells are stiffened with air? This material sample explores the possibilities of prototyping with airtight textile seams. instead of having to materials with highly contrasting qualities (hard/soft) connected in one material system, here one material can change between two states. Inflating the chambers will result in hard/stiff tiles, while deflation relieves the surface tension.
How does the surface curve when the cells are stiffened with air? This material sample explores the possibilities of prototyping with airtight textile seams. instead of having to materials with highly contrasting qualities (hard/soft) connected in one material system, here one material can change between two states. Inflating the chambers will result in hard/stiff tiles, while deflation relieves the surface tension.
30-03-20
II
Simulation of 3D Surface Tessellation
One goal of this research project is to find out how irregularity of tessellation influence the kinematics of a whole surface system. One way to approach the functionality of tessellation is to systematically model tessellated systems with parametric tools. These experiments show how the rules defining a flat surface are being applied to surfaces with positive or negative Gaussian curvature. The surface parcellation leads to the deformation of tesslleries that are projected onto parts with relatively high gaussian curvature. When the curved surface is being stretched as in fig.2 the triangular base grid is distorted resulting in the deformation of hexagonal cells. The more a surface is deformed/curved the higher will the degree of deformation for the affected hexagonal cell be. In the next experiment, we could evaluate the amount of curvature deformation going from 0-1 for each tessllerie, to evaluate the degree of distortion for each cell. Another way to go would be to define each hexagonal cell as undeformable but to move it according to the movement of the triangular base grid, which defines the center points of the cells. Therefore a normal vector relating to the curved surface has to be evaluated and later used to define the new position of each cell.
One goal of this research project is to find out how irregularity of tessellation influence the kinematics of a whole surface system. One way to approach the functionality of tessellation is to systematically model tessellated systems with parametric tools. These experiments show how the rules defining a flat surface are being applied to surfaces with positive or negative Gaussian curvature. The surface parcellation leads to the deformation of tesslleries that are projected onto parts with relatively high gaussian curvature. When the curved surface is being stretched as in fig.2 the triangular base grid is distorted resulting in the deformation of hexagonal cells. The more a surface is deformed/curved the higher will the degree of deformation for the affected hexagonal cell be. In the next experiment, we could evaluate the amount of curvature deformation going from 0-1 for each tessllerie, to evaluate the degree of distortion for each cell. Another way to go would be to define each hexagonal cell as undeformable but to move it according to the movement of the triangular base grid, which defines the center points of the cells. Therefore a normal vector relating to the curved surface has to be evaluated and later used to define the new position of each cell.
26-03-20
II
![]()
Physical Model approach towards Tessellated Systems
This is a first exploration of using a 3D Printer to simulate the relation of hard (tiling) and soft (interface) materials in one surface system. This low-fidelity prototype is created with a Zortrax 3D Printer extruding PLA on a TPU coated Tyvek fabric. The Tyvek is taped to the building plate which is preheated to 90 C°. The extruder produces a thin thread of PLA (0.4mm) which needs to be glued to the Tyvek. As the nozzle of the 3D printer heats up to 275 C° it easily burns the Tyvek, while when moving the nozzle to fast the extruded plastic (PLA) will harden to fast and not stick to the Tyvek. I found it profitable to reduce the building speed to 75% while keeping the extrusion speed at 90%.
This material test incorporates hard tiles onto a flexible interface. Interestingly the soft interface although being the minority material (in natural examples even 99% tiles to 1% interface) mostly predefines the kinematics of the surface. Whereas the tiles define some boundary conditions of extreme kinematics. It can also be observed that printing onto a flat interface allows no geometric curvature of the surface (Gaussian curvature). The interface only curves geometrically when stretching. Therefore more fabrics and net materials will be tested.
This is a first exploration of using a 3D Printer to simulate the relation of hard (tiling) and soft (interface) materials in one surface system. This low-fidelity prototype is created with a Zortrax 3D Printer extruding PLA on a TPU coated Tyvek fabric. The Tyvek is taped to the building plate which is preheated to 90 C°. The extruder produces a thin thread of PLA (0.4mm) which needs to be glued to the Tyvek. As the nozzle of the 3D printer heats up to 275 C° it easily burns the Tyvek, while when moving the nozzle to fast the extruded plastic (PLA) will harden to fast and not stick to the Tyvek. I found it profitable to reduce the building speed to 75% while keeping the extrusion speed at 90%.
This material test incorporates hard tiles onto a flexible interface. Interestingly the soft interface although being the minority material (in natural examples even 99% tiles to 1% interface) mostly predefines the kinematics of the surface. Whereas the tiles define some boundary conditions of extreme kinematics. It can also be observed that printing onto a flat interface allows no geometric curvature of the surface (Gaussian curvature). The interface only curves geometrically when stretching. Therefore more fabrics and net materials will be tested.
21-03-20
II
![]()
Meeting with Jacob Naumann - The randomization pattern
Jacob Naumann did his BA at MPI, where he was investigating the functionality of irregular surface parcellation. During his thesis, he developed a Grasshopper script allowing for gradually and parametrically controling the degree of “Voronocity”. Therefor a regular grid of points is constructed, around every point a circle is drawn. A vector (radius) from point to circle border is constructed, the angle in which this vector is constructed is assigned a random number between 0-360. Then each point is moved along the constructed vector, resulting in a more irregular pattern. This is a visualisation by Jacob Naumann.
Jacob Naumann did his BA at MPI, where he was investigating the functionality of irregular surface parcellation. During his thesis, he developed a Grasshopper script allowing for gradually and parametrically controling the degree of “Voronocity”. Therefor a regular grid of points is constructed, around every point a circle is drawn. A vector (radius) from point to circle border is constructed, the angle in which this vector is constructed is assigned a random number between 0-360. Then each point is moved along the constructed vector, resulting in a more irregular pattern. This is a visualisation by Jacob Naumann.
12-03-20 Input
Following the concept of Dual Tessellation described by Peter Pearce a hexagonal tessellation is always constructed around the center points of a triangular base grid. When employing Jacobs Grasshopper Script the regular order of the triangular base grid is distorted (as described above). Due to the duality of the base grid and the partitioning grid, the partition deforms accordingly to the deformation of the base grid.
Structure in Nature is a Strategy for Design
Regular Tessellation
A regular tessellation is a pattern of congruent regular polygons filling the hole plane, on which all vertices of the tessellation are surrounded alike. There are only three possible regular tessellations. They are tessellations of triangles, of squares and of hexagons. There are no other cases. This can be simply explained by pointing out that in order to subdivide the plane with polygons, the angles around each vertex must sum 360°. In the case of regular tessellations this means that only polygons can be used that have face angles that can be whole-number subdivisions of 360°. The triangle with face angles of 60° divides 360° into 6 parts. Therefor a tessellation of triangles has six polygons meeting at each vertex. A hexagon has 120° face angles, witch divides 360° into 3 parts, therefor a tessellation of hexagons will have three polygons meeting at each vertex. Less than three polygons meeting at each vertex cannot subdivide the plane. Therefore any polygon with face angles greater than 120° will not be capable of forming a regular tessellation.
Dual Tessellations
The concept to the reciprocal or dual network was important to our discussion of closest packed systems. As we shall see it is fundamental to. the understanding of the properties of all periodic spacial systems. We have already noted the duality of triangular networks and hexagonal networks. A dual network is formed by joining the centers of each polygon to all neighboring polygons through the shared edges. The dual network always forms polygons which are the domains of the vertices i.e. polygonal domains will have the same number of edges as there are edges meeting at the vertex it encloses.
Regular Tessellation
A regular tessellation is a pattern of congruent regular polygons filling the hole plane, on which all vertices of the tessellation are surrounded alike. There are only three possible regular tessellations. They are tessellations of triangles, of squares and of hexagons. There are no other cases. This can be simply explained by pointing out that in order to subdivide the plane with polygons, the angles around each vertex must sum 360°. In the case of regular tessellations this means that only polygons can be used that have face angles that can be whole-number subdivisions of 360°. The triangle with face angles of 60° divides 360° into 6 parts. Therefor a tessellation of triangles has six polygons meeting at each vertex. A hexagon has 120° face angles, witch divides 360° into 3 parts, therefor a tessellation of hexagons will have three polygons meeting at each vertex. Less than three polygons meeting at each vertex cannot subdivide the plane. Therefore any polygon with face angles greater than 120° will not be capable of forming a regular tessellation.
Dual Tessellations
The concept to the reciprocal or dual network was important to our discussion of closest packed systems. As we shall see it is fundamental to. the understanding of the properties of all periodic spacial systems. We have already noted the duality of triangular networks and hexagonal networks. A dual network is formed by joining the centers of each polygon to all neighboring polygons through the shared edges. The dual network always forms polygons which are the domains of the vertices i.e. polygonal domains will have the same number of edges as there are edges meeting at the vertex it encloses.
reading
Duality of base grid and surface parcellation
The exact amount of circles can be organized along a triangular grid of points and a rectangular grid of points. If we visually compare those two arrangements it is evident that the triangular grid of points can organize the circles more efficiently (bigger circles - tighter packed) without overlapping geometries.20-03-20 - virtual Prototype
![]()
II
![]()
When a Voronoi calculation is applied to the triangular grid of points we see that the regular base grid results in a regular parcellation of the surface. All tessleries have the same size. Once the row counts in Y-direction is manipulated the individual tiles deform regularly, from stretched hexagonal tiling to compressed hexagonal tilings. This is a demonstration of how controlling the points of a surface can help to control the parcellation of the surface.
Workshop
I joined an intensive parametric Design workshop lead by Wassef Dabboussi from LAVA (Laboratory for Visionary Architecture).
I joined an intensive parametric Design workshop lead by Wassef Dabboussi from LAVA (Laboratory for Visionary Architecture).
14/15-03-20 Input
Laser Cutting Test
Some of the generated patterns were materialized by laser cutting the pattern into cardboard.
Some of the generated patterns were materialized by laser cutting the pattern into cardboard.
03-03-20 - physical prototype
Grasshopper Exploration
A Grasshopper script that controls a triangular grid of hexagons. An underlying circle influences the size of each hexagon. The hexagons nearer to the circle will be influenced stronger. I tested a morphological space ranging from 0.1 radius to 1.0 radius was tested.
A Grasshopper script that controls a triangular grid of hexagons. An underlying circle influences the size of each hexagon. The hexagons nearer to the circle will be influenced stronger. I tested a morphological space ranging from 0.1 radius to 1.0 radius was tested.
24-02-20 - virtual Prototype
![]()
![]()
Structure in Nature is a Strategy for Design
Tessellated material systems can be described, evaluated and analyzed by geometries. In “Structure in Nature Is a Strategy for Design” Peter Pearce defines underlying geometrical parameters to describe biological structures. Closely analyzing the principles of biological structures and employing these as inspirational strategies for structural and aesthetic design. Using these parameters, he intensively explores geometries and structures, that were not yet found in natural systems. Pearce suggests the following ways of analyzing structures:
The principle of closest packaging (Triangulation): If circles are tightly packed, as densely as possible, and their centers are joined, an array of triangles results. This base grid defines how the circles are packed, a triangular grid can pack circles more efficiently than a rectangular grid.
Partitioning and Tessellation: Let's say a surface is partitioned by circles. In the case of closest packed circles (triangular base grid), small concave triangles are formed between the circles. Although the circle is one very economic way of closing a surface, it can be partitioned more efficiently, if we allow the circles to transform into hexagons. This becomes the most economical method for partitioning a surface into equal parts.
The packing of hexagons reveals the fundamental relationship of the triangular order of close packed circles with the requirements of minimal partitioning.
Tessellated material systems can be described, evaluated and analyzed by geometries. In “Structure in Nature Is a Strategy for Design” Peter Pearce defines underlying geometrical parameters to describe biological structures. Closely analyzing the principles of biological structures and employing these as inspirational strategies for structural and aesthetic design. Using these parameters, he intensively explores geometries and structures, that were not yet found in natural systems. Pearce suggests the following ways of analyzing structures:
The principle of closest packaging (Triangulation): If circles are tightly packed, as densely as possible, and their centers are joined, an array of triangles results. This base grid defines how the circles are packed, a triangular grid can pack circles more efficiently than a rectangular grid.
Partitioning and Tessellation: Let's say a surface is partitioned by circles. In the case of closest packed circles (triangular base grid), small concave triangles are formed between the circles. Although the circle is one very economic way of closing a surface, it can be partitioned more efficiently, if we allow the circles to transform into hexagons. This becomes the most economical method for partitioning a surface into equal parts.
The packing of hexagons reveals the fundamental relationship of the triangular order of close packed circles with the requirements of minimal partitioning.
12-02-20 - reading
Why the seahorse tail is square
Michael Porter et al. present a prototype-based approach of investigating animal behaviour and locomotion. They use 3D printing and CAD Software to explore the mechanics of the seahorse tail. Employing tools commonly used in design disciplines allows them to build functional models of hypothetical tail systems. These can be compared to reproductions of common tail systems. In this paper, the articulated outer shape of the tail is explored. As cylindrical seahorse tails do not exist in nature, 3D printing allows building functional models of this hypothetical system allowing to investigate what mechanical advantages an articulated square prism may have over a cylinder one. This approach was found to be particularly useful to examine the mechanics of systems in vertebrates that are difficult, impossible to study in nature. Design strategies are employed to generate alternative shapes that can be contrasted with the existing ones in nature. With scientific measures of evaluation, this method can help to map parameters that reveal the functionality of natural structures.
Michael Porter et al. present a prototype-based approach of investigating animal behaviour and locomotion. They use 3D printing and CAD Software to explore the mechanics of the seahorse tail. Employing tools commonly used in design disciplines allows them to build functional models of hypothetical tail systems. These can be compared to reproductions of common tail systems. In this paper, the articulated outer shape of the tail is explored. As cylindrical seahorse tails do not exist in nature, 3D printing allows building functional models of this hypothetical system allowing to investigate what mechanical advantages an articulated square prism may have over a cylinder one. This approach was found to be particularly useful to examine the mechanics of systems in vertebrates that are difficult, impossible to study in nature. Design strategies are employed to generate alternative shapes that can be contrasted with the existing ones in nature. With scientific measures of evaluation, this method can help to map parameters that reveal the functionality of natural structures.
05-02-20 - reading
![]()
![]()
