Blossoms by Merate Barakat

Blossoms Sonic

Response

Studio

Sarah Winkler Steven Arave

Table of Contents 3 5 6 7 9 14 15 18 19 22 24 29 30 31 35 45 50 53 57

Introduction 8.28.13 8.30.13 9.02.13 9.04.13 9.06.13 9.11.13 9.13.13 9.14.13 9.17.13 9.24.13 9.29.13 10.01.13 10.07.13 Final Boards Final Presentation Appendix 1-2 Arduino Sketch Appendix 3-5 Grasshopper Definition Appendix 6- Photos Appendix 7- Videos

The physical environment is not only experienced through visual sense. We have been encouraged in this studio to consider that architectural design is found to be successful when it manipulates one or more perceptions in order to convey a certain aesthetic significance. We decided to manipulate human infrared energy to create a sculptural choreography. The captured human energy triggers motion in the opening and closing of the blossoms which in turn layers with a simple tone to create a visual, aural, and human responsive as one intertwined paradigm. The layered and open structural framework reflects this intertwining and encourages the visitor to step closer to see whatâ&#x20AC;&#x2122;s going on and interact further.

Folding Wall Screen

8.28.13 We began today with a brainstorming session. We individually researched different kinetic precedence online. We then collaborated and discussed the pros and cons of our ideation. We came up with three options to present to Merate. 1- 2- 3-

“Coiling Pedestrian Bridge” “Variation of BMW exhibit” “Light Wall”

Based on our discussion with Merate we discovered numerous issues with options one and three. We then further brainstormed and discovered that we could design a compelling exhibit using the BMW exhibit as precedence. With the BMW exhibit in mind, our discussions became increasingly more creative. We began to discuss possible materials, sound responsiveness, sound creation, colors, lighting variations, and structure. We wanted to maintain the spirit of the BMW exhibit, but also distance ourselves from a literal translation. We discussed lighting options, cloth, fabrics and beads. We ended the session by deciding to further explore suspended beads as a medium.

Open Position

Coiling Pedestrian Bridge Closed Position

Variation of BMW exhibit

9.02.13

8.30.13 We are starting to move away from just looking at examples of cool kinetic sculptural installations on-line and think about the steps that we need to take to develop our own project. Some concerns that we have discussed are of course schedule and budget. These are issues for all projects. One big difference for this studio project is that we anticipate incorporating software and hardware that we are not yet up to speed with and don’t know how decisions made today will effect budget and schedule as the project develops. We have been thinking that it can cause two extremes. (1) we limit our designs to ensure that what we picture today could be built based on minimal knowledge of systems we have now, or (2) we surge forward and blindly trust that we have the ability to solve any problem as it arises while maintaining schedule and budget. The path forward is probably somewhere in between.

The chip board stiched-together component idea became frustrating and did not look practical or economical in terms of time. The aesthetic cried out messy versus anything else that might have been created with the stitch work added to angled elements.

Stitching Component

Worked on an alternative method of fabricating the component, at least for the purposes of the sketch up model. If the component is made of folded paper with a mechanism at the underside of the ridge folds that would stiffen and allow for “telescoping” action that might simplify things. The connection where six components meet needed to be thought through. A trip to Ace Hardware…found aluminum tubes in varying small diameters that could be used to develop telescoping mechanism. The larger tube could be glued to the underside of the ridge fold (3 of these per component.)

Today we are developing at the component level and coming to an agreement on what the deliverables for the pin-up on Wednesday will be and methods of construction. We decided it was important for our project to have the ability to develop into a building façade in the future. In a future development the façade could be programmed to respond to temperature versus sound.

The shop was closed so most of the remainder of the work session was on 3-d modeling designing the connection, designing the “sketch model” and making a materials list.

We’ve agreed to work on a wall that responds to sound and could potentially create sound also. The wall that we are considering now would be made up of interlocking equilateral triangular components. The triangular component would be designed to fold in on itself. The folding could be programmed and mechanized with hardware that can push/pull in linear action….similar to how umbrella operates. Both of us built the component in Rhino and played with different positions of open and close so that we could understand how the component would move on a very basic level.

Sketch Model

Component Sequence

Component Pattern

Sonic Responsive Design Studio

Sarah Winkler. Steven Arave.

Pin up One

9.04.13

Component Proposal

Today we had our first review for our sonic wall. We understood that we should work on the following: • Consider the material of component construction. Suggestion was to use a foam sheet that can be cut in the laser cutter. In discussing later, we both really like the texture of paper and the possibilities for light to transfer through.

Component Aperture Sequence

Component Pattern

Component Connector

Connector Detail

Sketch Model Side View

Sketch Model Top View

Wall Visualization

Wall with Scale Figure

• We considered incorporating a rotational aspect to our wall. After discussion we concluded that a simple blooming motion was the direction we wanted to pursue. Most likely eliminating the corner supports and really emulating the movement and support of the “cocktail umbrella.” • Add gear mechanism to control movement, the more clever the gear system is designed, the number of motors and aduino boards goes down…that’s the goal. • It is not likely footsteps would emit enough energy to produce a response. Suggestion was that small microphones be incorporated that could be installed around 5’ and 4’ height that would allow clusters of components to respond to voice activation. In our meeting later we decided to keep things simple and not try to produce sound as a secondary product of the movement.

Fall 2013

September 4, 2013

9.06.13 3. Linear motion is achieved with a horizontal component which in elevation view would have teeth designed to engage the driving gear described in #2. Range of motion of this linear action is determined by analyzing the 1:1 scaled component and determining how much distance is needed to bring the component into the open position. This distance also affects the circumferential distance that is needed at the original driving gear assembly that changes direction described in #1 which is then related to the diameter for that assembly. 4. At this point we think that any variability in opening and closing the component is achieved through changing the speed of the push-pull movement. Changing speed is accomplished by changing the diameter of the gear that drives the horizontal component with teeth 5. This gear cluster does not float in space so adequate supports need to be designed that allow for rotation or slipping through without to much obstruction from friction. 6. Right now it seems like to get any kind of variance in the opening and closing, each component that is going to be moved needs to have a gear cluster. Those can be tied together linearly with dowels. Hopefully some linear rows can be grouped to further minimize motor controls. How much tying together of gear clusters can be accomplished is likely a function of weight of materials, capacity of motors, “etc.” “Etc.” in this case means there is likely a lot of trouble shooting and experimenting ahead of us because we haven’t worked this project from both ends. Knowing more about capabilities and limitations of motors and motor controllers, “etc.” along with their associated costs would likely result in a more clever and elegant solution. 7. Materials and methods of fabrication of the gear clusters has been considered. For preliminary models will likely cut gears out of 1/8” MDF with the laser cutter. For such small components the laser cutter seems to cause the cut piece to slip around a bit and for gears where we imaging some accuracy is required for the teeth to engage and not bind that could be a mess. Another thought was to 3d print but we are not yet at a point where the gear file can be presented to Plastik 8. Banana for evaluation as to feasibility and cost for 3d printing.

We have two main objectives to get us to the next phase. A. First is to refine the design of the component to: 1. Simplify the mechanics, 2. Minimize connections, 3. Evaluate different weights of paper (we are still interested in using paper versus foam, the aesthetics of paper seem to be more in line with our vision) 4. Design stiffner/telescoping capability; looking for something light weight and with right amount of flexibility to form hinge joint and to wrap around dowel that will have the push-pull motion. B. Second is to design “gear clusters” that will have the following capabilities: 1. End result is a push and then pull linear movement of component. At this point it looks like it needs to be made up of a main driver gear that will drive another gear in two directions. Theoretically that can be accomplished with a gear that has teeth on the outside radius of a smaller gear for a certain percentage of its circumference and teeth on the inside radius of a larger gear for a percentage of its circumference (related of course to the smaller gear’s percentage). Smaller and larger gear are fixed to each other and operate therefore operate with same center of rotation. Driven gear is stays at one location with respect to the driving gears and has a diameter approximately equal to the difference between the larger and the smaller driving gear diameter. The driven gear would hopefully move in one direction when teeth are engaged with the smaller driving gear’s teeth and switch rotation direction when it runs out of teeth on smaller gear and picks up teeth on larger diameter gear….theoretically. 2. The driven gear would in turn drive another gear that could have variable diameters that would drive the linear motion.

Component Models

Flow Chart

Gear Component

Gear Pattern

Wall Pattern

Gear Component

Gear Pattern

Wall Pattern Wall with Scale Figure

Wall with Scale Figure

9.11.13

Blossoms Sonic Response Studio

Pin Up Two

The past week we have been focusing on the component schemes and design of the gears. After speaking with Merate we began to question the validity of the gears and mechanical movement of the wall system. We were directed to look into a shape metal alloy which could reduce the number and complexity of our gear system. We researched different metal alloy systems and settled on exploring a shape metal spring. This spring would contract and expand with heat or electricity and push/pull our component open and closed. This operation would reduce the need for gears and complex mechanical operations. We have ordered a sample of the metal to explore its attributes and potential to be incorporated into our project. While this operation could potentially simplify the movements it requires electrical wiring which we would need to learn.

An in depth exploration in the component design, assembly, gears and wall composition. We have made significant progress in design composition, materials, gear functions and efficiency. Modifications are still needed to gears and component resiliency. Gears and components need to be assessed and modified to withstand repetitive use.

We want to keep our options open, therefore we are not abandoning our gears that have been developing. However, we are exploring the possibilities of simplifying our mechanical system. Ideally we would reduce the number of gears. Currently we have all wall parts moving independent of each other. If we continue to pursue a similar track we need to explore moving multiple components with limited gears.

Wall Proposal

Gear Diagram

Gear Sequence

Wall Proposal

After further brainstorming we have decided to continue on a path with developing our â&#x20AC;&#x153;blossomsâ&#x20AC;? while simplifying our mechanics. We will explore the metal spring in addition to designing a simple solution to gears.

Wall Proposal with figures

Fall 2013

Gear Components with figures

Sarah Winkler. Steven Arave

9.13.13 • Wood dowels purchased from U bookstore and used for shafts are not straight. Even though labeled with same diameter they actually vary in size.

Pin up two We constructed a panel of 21 gear driven components.. The component sizes are 1:1. It was a good choice to go 1:1 on component size because it allowed us to see the performance of component and gearing mechanism and to identify where the problems would occur versus waiting until later date to scale up.

• The large driving gear which causes the bi directional movement is made up of three parts that needed to be carefully aligned. There is a center section with teeth on the outside radius (approx 2/3 rds around) and around about 1/3 of the inside diameter of the exterior ring.

The gear clusters were fabricated using laser cut MDF (1/8”). Each gear cluster consisted of the following pieces: Base plate, layer 1 1. 2. Base plate, layer 2 with ridges for linear track and pockets for shaft base supports 3. Larger driving gear producing clockwise and counter clockwise rotation in driven gear 4. Driven gear 1 5. Driven gear 2 is driven by gear 1 and is located over the linear track 6. Linear track that is designed to move backwards and forward 7. Miscellaneous dowel shafts As soon as the fabrication and assembly process began, problems started to surface. • The time it took to laser cut all the parts and pieces for 21 gear clusters was approximately 5 hours. The full scale installation is likely to have 162 gear clusters….which would be around 40 hours of laser cutting. Laser cutting MDF is not a practical fabrication method based on available time alone. • The laser cut does not result in a straight edge (perpendicular to the surface being cut) so small gears cut like this don’t mesh well and/or cause binding. For an functional result precision is cutting, for this reason laser cutting MDF is not acceptable. • We should have considered spacers and bearings made out of material with lower coefficient of friction. These types of fittings weren’t considered in the design but they actually would have helped the situation a lot.

• Differences in diameter of axels and holes laser cut in gears caused the gears to angle a bit which causes wobbly motion at best in the gears. • Installing the gears clusters on the back of the board in pockets, aligning the gear clusters in each row and finetuning the location and straightness of the gears was a messy operation totally lacking in precision or control. • The backboard itself has deflection. In this case it was just two layers of 1/8” MDF and about 2’ x 3’…the full scale would be constructed differently • The accumulated effect of all the above resulted in a dysfunctional gear system. All the parts and pieces were there but there was too much binding up of the gears to be able to utilize. It’s not an understatement to say that precision is critical for geared systems to work.

Blossoms

Gears

• Assembling the gear clusters with the component on the front of the wall led to another discovery that the travel distance on the gear system was greater than the travel distance allowed by the components frame system…so without changing the circumference distance with teeth on the driving gear, they system would likely push a hole through the component. In the review session, it was recommended that we drop the gear approach and follow up with the shaped metal alloy. This would likely be installed in a spring configuration. When heated (by electricity) then it would straighten out. Straightening out would cause the opening of the component.

9.14.13 We’ve been advised that the paper components are not the best choice because they seem too stiff and that we should look into fabrics. For Monday, we are going to make another stab at a single component model that incorporates the fabric versus paper, and shape memory alloy to produce the linear motion.

Initially we used thin paper in the blossom component. After unsuccessful attempts, we decided to use a heavier and textured paper. In the beginning attempts were not promising due to instability with the chip board. After working further and creating several types of models, we came up with a component design that would both contract and retract more easily.

We’ve purchased some samples of fabrics in two general categories. First are fabrics that stretch in both directions….”4 way stretch”. The second category are fabrics that don’t stretch. The fabrics that don’t stretch would require a stretchy connection at each of the three points of the component. The Dupoini silk (not stretchy) has the best look and feel in terms of replicating what we both liked about the paper. If it turns out we need to add some stiffness to the fabric we can get some iron on interface fabric. Grommets were purchased because it sounded like something we would want to have to keep the fabric from tearing at connection points.

During out presentation, we were highly discouraged from using paper as the folding element. Using a fabric, nylon or craft paper would give the wall a more elegant, flexible, and light. More flexibility would allow the contractions to be more seamless and effortless. Lighting the weight would also reduce the strain on the gear or spring function. For this next week, we will focus on using a different material for the folding portion. Also the structural material will likely need to change to support the new material and give it a uniform look. Due to the material changes the structure of the blossom will also need adjusting to support the slightly altered movement and structure.

Another trip to ACE hardware was made to get parts and pieces for the connection of the triangle corners to each other and a fixed distance from the back board. 1/8” threaded rod with washers and nuts was selected at least for this model. Tomorrow we’ve scheduled a trip to a hobby store that we’ve just heard about to get ideas for materials for connections and for the “umbrella” framework. They may also have shape memory alloy actuators. In the meantime, we are doing research into sensors (sound, motion, light), arduino kit components to get a better idea of where we’re going with the final installation in terms of response mechanism, control and electrical wiring.

Blossoms

Wall Section

9.17.13 Shape memory alloy has the ability to only contract when a current is applied. This can create a pulling force. The contraction is only a percentage of the length. I evaluated Images Scientific Instruments nitinol alloy products. They had the most information from which to evaluate on line. Also, made a phone call to verify my understanding of how nitinol works with respect to creating a push-pull action. Their products are1:

Diameter Size (Inches)

Flexinol Wire: Flexinol wire is a trademark for trained nitinol actuator wire in a variety of diameters from .002 to .015. The wires contract (typically 2% to 5% of their length) like muscles when electrically driven or heated. The advantage of this wire is that the small size, light weight, ease of use and silent operation allow it to replace small motors or solenoids. However to get 1.67” travel distance then would need 33.4” to 83.5” of wire. A schematic of potential way this might work is in the plan view sketch 1. It’s kind of crazy but could add some sort of 3’d element in the back of the wall with the stretching of wires. Each row could be controlled individually, but not each component. Estimated cost for wire and the compression springs that would be required is:

Approximate* Current Off Time 70° C Off Time 90° C Resistance Maximum Pull Contraction* at Room Temperature LT Wire** HT Wire** (Ohms/Inch) Force (grams) Time (seconds) (mA) (seconds) (seconds)

0.002

12.0

0.3

0.1

0.003

5.0

100

0.5

0.2

0.004

3.0

150

180

0.8

0.4

0.005

1.8

230

250

1.6

0.9

0.006

1.3

330

400

2.0

1.2

0.008

0.8

590

610

3.5

2.2

0.010

0.5

930

1000

5.5

3.5

0.012

0.33

1250

1750

8.0

6.0

0.015

0.2

2000

2750

13.0

10.0

BioMetal Fiber: Biometal fiber is .006” dia x 39” long. It is a fiber-like actuator designed to contract and extend like muscles. It is soft and pliable like a nylon thread under normal conditions, becomes stiff like piano wire and sharply contracts when a current is fed through it. If the passage of a current is stopped, it will soften and extend to its original length. It can produce 150 grams of force (5.2 lbs); typical cycle rate at 20 degrees C is one minute. This would be used as an individual component control. Assume 50 out of 100 components are active, four components could be activated from one 39” long fiber. The cost for 50 components would be $586, for 100 components it would be $1,173. However this does not include the crimps that would be needed on each end to connect to….they are about $0.25/ea. For reliability these should be added at the manufacturer’s which is an additional $30/set up fee.

• 93” per row = 7.75’ @ $2.50/per foot x 20 rows = $387.5 for wire (.004 dia selected for cycle time and produces about 5.2 lbs of force) • 100 compression springs (model selected has a rate of 3.4 lbs/in which could produce 1.53” of travel. Cost is $2.30/ea. For 100 springs cost is $230

BioMetal Helix; The advantage of the BioMetal Helix over standard nitinol wire is that it may be stretched and contracted to twice its length. Could produce 1.5” of movement. Typical cycle rate at 20 degrees C is one minute. At $46.95 it would need 20 if controlling each row….$939 Nitinol Expansion Spring; could work but is $44.95/ea, would need 20 if controlling each row …$900.

Total estimated cost of just nitinol wire and compression springs is $617

Reference 1 Images Scientific Instruments website http://www.imagesco.com/nitinol/nitinol-index.html accessed 9-17-2013.

Advantages versus Disadvantages of Nitinol shape memory alloy for this application

We have made significant progress since the last journal entry. Previously, we had been proposing a solid wall between the blossoms and the motors. We were encouraged to think about altering the overall composition of the wall. We have since significantly altered the form of our wall. The wall is a horizontal orientation with openings and flowing form. The result is a more interesting, dynamic, and lighter. Structure, motors and shafts will be exposed as a result of the form.

Advantages: • Light weight • Small • Silent • could look cool Disadvantages: • expensive $$$ • not available locally for purchase as far as we can identify; so there is a significant problem for experimenting and validating scheme. The order takes processing and shipping and even with expedited shipping takes four to five work days. • the flexinol wire has information on electrical resistance and approximate current at room temperature but we are not knowledgeable enough to know whether the selection is correct and what it means for hardware requirements downstream to power this up • the nitinol product that we received as a sample heats up significantly when the current is on. Really have to consider whether or not this is a fire hazard. The Flexinol wire is the least costly of the nitinol options but each row • would be doing the same thing, no individual control or variation. • This is an evaluation based on a few days of research on-line. In order verify that the proposed schemes are feasible, new materials would need to be obtained immediately. These would not be received until Saturday based on the experimental wire order. To test the reaction to current, we would need to have the electrical skills to hook up to power. The 9 volt battery used on the experimental wire either burned out or the wire couldn’t handle the back

Blossoms

The blossoms will now move as a group of six components rather than independently of each other. This does several things for our design. First, it reduces the cost in motors and materials. This move also further emphasizes our blossoms blooming as now six will contract and expand as a larger blossom. Our structure will appear light and frames negative spaces. Metal or wood will be used to support both the blossoms and the motor platforms. We will explore using Fimo clay as the connection joint for the frames that tie the components. Fimo clay is moldable until baked where it becomes solid. If Fimo clay is not successful alternative adhesives could include soldering metal or glue.

Sarah Winkler. Steven Arave

Concerns: time available to complete project and cost of project. The costs estimated in this report of findings are only for the linear actuator. The costs including additional development models are likely to put this studio’s costs over $1,000 as a very rough estimate..

9.24.13

We will group six components together into a “cluster”. The scheme we’ve agreed upon distributes 11 clusters across the wall. To save cost and because of limited time we are talking about this wall being a partial height that is anchored to a base, that sits on a table. The clusters would be located from 36” to 76” above the floor. Each component within a cluster would move by means of an individual push rod, but the six push rods would be tied together with a light weight frame so that only one mechanism is needed to activated all six components. We want the wall be activated by motion….so when a person comes up and passes within range or stands and waves their arms. Each cluster would be equipped with one sensor at the center which in turn controls the servo motor. The servo motor would be programmed to rotate within a 140 degree arc which is calculated to give a 1.67” travel distance. The 1.67” travel distance is the distance that the component push rod needs to travel to go from the starting position which is slightly open to the fully open position. We decided it may not be a good idea to connect the cluster push rod to the servo motor directly. The rod would take a bit of a rotational movement that would very likely add extra torque requirements to the motor and probably end up bending all the push rods. To make the travel purely linear, some a vertical gear needs to be anchored to the servo motor which in turn needs to be anchored sideways to a vertical support. The gear is sized to accommodate the 1.67” of movement and drives a linear piece with teeth on top. The push rod for the cluster would be anchored to this linear motion piece. For the mid review on Friday, this gear system will be fabricated again from laser cut MDF. It’s a much simpler gear system than attempted before so maybe there will be exponentially less problems.

We developed a grasshopper definition to simulate our wall. Starting with a surface in Rhino that represented one half of one third of a single component, we were able to first built a component that would open and close using a number slider. From there the definition constructs a cluster. Since the cluster is designed to be operated by a single mechanism, all components within a cluster have the same angle of opening. Next the cluster was applied to a triangular grid. We had decided on the placement of the 11 clusters relative to each other a couple of days ago by a manual method. Looking at many variations and then agreeing on which one best fit with our vision of the wall. Another way to select the placement (which may be pursued if time allows) is to use Galapagos in Grasshopper along with fitness criteria to narrow in on the distribution. However, for this definition, a process of culling points from surfaces created on the triangular grid was used to narrow in on the center point placement of the 11 clusters. Clusters were moved to the new locations. An attractor point representing a person moving in front of the wall was included. The input to each clusters which controls the open angle of the individual components was modified to be a function of the distance between the attractor point and the center of the cluster. The attractor point is set up on a straight line (curve) so that it can be moved around to simulate the response of the wall as someone moves past. The clusters closest to the attractor point have the greatest degree of openness and the further away they have decreasing angle of openness.

Wall composition

We are also working on developing the electrical components parts list for the wall. We’re thinking that the best way to work the wall is that each cluster has a proximity sensor installed at the center. Right now we think the best way to set up the system to allow for individual function and operational reliability is to connect 11 micro controllers to the Arduino board. Each micro controller would control one sensor and corresponding servo motor.

Interim Model

The selection of the servo motor needs to be revisited once we understand the torque requirements. The physical model we are preparing for Friday will help with this determination…its not put together yet. Technically, some sort of torque gage needs to be used to make the selection most accurately.

The Arduino Micro packs all of the power of the Arduino Leonardo in a 1.9” x 0.7” (48mm x 18mm) size. Although it may look like a stick of gum, its actually a USB-native 8-bit microcontroller, with 32K of flash, and 2.5K of RAM. You can program it directly via the USB micro connector on one end (or for the advanced users, the 6-pin ISP header). The Arduino Micro is a microcontroller board based on the ATmega32u4, developed in conjunction with Arduino.cc and Adafruit.

The following is our parts list of the main electrical components (not including wires, cables, resistors, jumpers, and any other required part of a functional electrical system):

Onboard is the processor, crystal, micro USB connection with fuse, Reset button, ISP low-level reprogramming header, ON LED, TX and RX LEDs, a extra user pin #13 LED, and a 3.3V regulator. You’ll probably also want to pick up a micro-B USB cable to connect this board to your computer

We already have the Arduino Board. • Micro Controller (11 ea) available from Adafruit http://www. adafruit.com/products/1500 $22.95 ea

You can use it for basic Arduino sketches as well as USB client projects like making it appear as a mouse or keyboard to your computer! Although it is not shield compatible - it does have all the same pins as a Leonardo. Its best for when you want to shrink your project down or use it on a breadboard. This version of the Micro comes without headers pre-soldered on and runs at 5V logic with a 16MHz crystal.

Gear system

Interim Model

• Proximity sensor (11 ea): available on Amazon, $5.98 ea • Battery packs (22 ea??): available from Adafruit, $2.95 ea http://www.adafruit.com/products/830 not sure how many we need…might need more to power sensors?

Motion sensor

Battery Pack

9.29.13 We were as well prepared for the Interim Review as possible. The prototype cluster did not have a motor attached because we are still climbing a steep learning curve prior to ordering all the appropriate mechanical and electrical components. We did have an arduino uno, servo motor and PIR sensor set up and running successfully. We also had a few grasshopper simulations that were activated by a sensor to present (single cluster, wall with eleven clusters, and cluster prototype with moving gears illustrating opening and closing).

We were as well prepared for the Interim Review as possible. The prototype cluster did not have a motor attached because we are still climbing a steep learning curve prior to ordering all the appropriate mechanical and electrical components. We did have an arduino uno, servo motor and PIR sensor set up and running successfully. We also had a few grasshopper simulations that were activated by a sensor to present (single cluster, wall with eleven clusters, and cluster prototype with moving gears illustrating opening and closing).

Two key points were brought up in the Interim Review. The first point being that our installation doesnâ&#x20AC;&#x2122;t create a compelling sound. Originally we had planned on either digitally creating a noise or using the folding of the blossoms as a sound. Unfortunately the silk makes little to no noise. Earlier we had been encouraged rather to embrace the sound of the servo motors running when this had been discussed in studio. The jury was not convinced that the motors were enough sound to make a compelling argument for a sonic studio.

The second point was concerning the structure surrounding the blossom chain. For our prototype cluster model we just had a vertical dowel supporting the cluster knowing that we needed to think through an appropriate system to brace the wall and support mechanical elements of the installation for the final. The jurors encouraged us to consider using the same open frame elements that create the cluster to provide the bracing and support to base. We are moving ahead using the hexagonal shape as the supporting element to the exhibit. We can then use the additional supporting elements to run wiring along, rather than just hanging the wires haphazardly. The additional supports will also help create a more stable exhibit. We were encouraged by a few good comments during the jury. The jurors like our form, presentation and the overall concept. It is now important for us not to lose sight of these and other good points to the project. We will continue to work on our presentation flow to support our concepts.

10.01.13 high torque metal geared ball bearing servo, featuring an all metal gear train and 2 ball bearings on in the gear train for added smoothness, strength and precision. This servo is o-ring sealed and waterproof.

As time quickly passes, we need to produce the elements for the final exhibit. We can start on the blossom component so they can be incorporated into the overall piece later. Taking time to carefully produce these elements is needed to avoid fraying of the silk. Significant work is needed on the mechanics and electrical element of the project. We are researching different types of servo motors along with needed gearing, joints, etc. to achieve linear motion, rather than the rotational movement that a conventional servo motor produces. We have code that has successfully gotten the PIR sensor to respond and initiate a servo motor. There is not sufficient time remaining for experimentation and dealing with our knowledge dysfunction. We are going forward with electrical and mechanical components that may seem redundant or unnecessarily expensive route to go for those with greater knowledge. However, we have a deadline and we need a system that is as reliable as possible that we can put together with our team’s current level of knowledge if we are to take this project seriously.

Specs: Voltage: 4.5-6V Speed: 0.26sec/60deg(4.8v) 0.22sec/60deg (6.0v) Torque: 10kg.cm (4.8v) 12.8kg.cm (6.0v) (138 oz-in = 8.6 in-lbs @4.8v) Size: 40.9mm x 20mm x37.75 mm Weight: 58gram Motor: Brushed Gear Train: Metal geared Ball Bearing: 2 Type: analogue

Weight (g) 58 Torque (kg) 12.8 Speed (Sec/60deg) 0.22 A(mm) 45 B(mm) 41 C(mm) 39 D(mm) 20 E(mm) 53 F(mm) 30

Additional work is need on the structure and form of the final model. Supports are needed to resist rotational movements and deflection. Electrical components will be housed in the continuous base that supports the wall at 30” above the floor. We visualize the final installation to set on a table with approximate dimensions of 8’ x 3.5’ and a depth of 8” to 12” in the wall. Base depth is likely to be 4”, with overall plan dimensions to be determined once the wall structure is complete. This project will need lots of trouble shooting time so we can only make forward progress. Up to now we have made two steps back for every step forward. All elements of the project need to be in place by Wednesday evening in order to allow for sufficient trouble shooting and any minor modifications. We have made a big effort to stay ahead on all aspects

Final Presentation 10.11.13

10.07.13 Materials List for Final Assembly • • • • • • • • • • • • • • • • • • •

white Dupuioni silk, 50” wide x 3 yards medium weight fusible interface material, 20” wide x 5 yards 5/32” x 36”, diameter aluminum tube, 44 ea 3/32” x 36”, diameter aluminum tube, 10 ea 4-40 push rods with ball joint, 11 ea Medium nylon pivot pins, 396 ea Arduino micro boards, 11 ea bread board, 11 ea PIR sensors, 11 ea piezo, 11 ea standard high torque servo motors, metal gears, 11 ea 6 volt battery packs, 11 ea AA batteries, 44 ea 9 volt battery clips, 11 ea 9 volt batteries, 11 ea wire, 22 AWG, red, black, white, 80 lf each color 1/8” x 2’ x 4’, MDF panels, 6 ea ¼” x 2’ x 4’, MDF panels, 16 ea 2” x 4” x 8’ boards, 4 ea

Blossoms Sonic

Response

Studio

Problem We were challenged with designing a kinetic sculptural installation using processing and parametric features. The exhibition will be interactive with the user. We have embraced the sound produced by the motors as the sonic aspect of the installation. We believed it was critical that the installation could have an architectural application in the future, potentially activated by solar radiation rather than motion.

Ideation

Process

Component Sequence

First Blossom Pattern

First Wall Composition

Second Wall Composition

Fall 2013

Interim

Midpoint

Process Diagram

Front

Gear Detail

Interim Gear

Rear View

Front View

Model Rendering

ARCH

6005

Final Product

Open

Closed

Mechanical Diagram, Plan View

ProďŹ le

Sarah Winkler Steven Arave

Blossoms Sonic

Response

Studio

Sarah Winkler Steven Arave

equilateral.

cluster.

wall.

kinetics.

fabrication.

equilateral.

cluster.

wall.

kinetics.

fabrication.

Midpoint

Front

equilateral.

cluster.

wall.

kinetics.

fabrication.

equilateral.

cluster.

wall.

kinetics.

fabrication.

equilateral.

cluster.

wall.

kinetics.

fabrication.

equilateral.

cluster.

wall.

kinetics.

fabrication.

equilateral.

cluster.

wall.

kinetics.

fabrication.

Open

Closed

equilateral.

cluster.

wall.

kinetics.

fabrication.

equilateral.

cluster.

wall.

kinetics.

fabrication.

Blossoms

Sonic Response Studio

1 //////////////////////////////////////////////////////////////////////////////////////// void move() { // Sound on for 1.5 seconds tone(speakerPin, NOTE_C4, 1500);

//PIR Motion triggered servo movement #include â&#x20AC;&#x153;pitches.hâ&#x20AC;? #include <Servo.h> // pin variables const int speakerPin = 8; Servo myservo; const int sensorPin = 12; const int LED = 13;

// move gear backward for(pos = 0; pos < 120; pos += 1) { myservo.write(pos); delay(40); }

// servo position variable int pos = 0;

// move gear forward for(pos = 120; pos>=1; pos-=1) { myservo.write(pos); delay(40); }

//////////////////////////////////////////////////////////////////////////////////////// void setup() { // set up pins and modes myservo.attach(4); pinMode(sensorPin, INPUT); pinMode(LED, OUTPUT); digitalWrite(LED, LOW);

}

// reset servo myservo.write(pos); delay(1500); } //////////////////////////////////////////////////////////////////////////////////////// void loop() { // if the sensor send a signal detecting movement... if(digitalRead(sensorPin) == HIGH) { // move servo move(); // delay for 1.5 seconds delay(1500); } }

2 //Pitches H /************************************************* * Public Constants *************************************************/ #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define

NOTE_B0 31 NOTE_C1 33 NOTE_CS1 35 NOTE_D1 37 NOTE_DS1 39 NOTE_E1 41 NOTE_F1 44 NOTE_FS1 46 NOTE_G1 49 NOTE_GS1 52 NOTE_A1 55 NOTE_AS1 58 NOTE_B1 62 NOTE_C2 65 NOTE_CS2 69 NOTE_D2 73 NOTE_DS2 78 NOTE_E2 82 NOTE_F2 87 NOTE_FS2 93 NOTE_G2 98 NOTE_GS2 104 NOTE_A2 110 NOTE_AS2 117 NOTE_B2 123 NOTE_C3 131 NOTE_CS3 139 NOTE_D3 147 NOTE_DS3 156 NOTE_E3 165 NOTE_F3 175 NOTE_FS3 185

#define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define

NOTE_G3 196 NOTE_GS3 208 NOTE_A3 220 NOTE_AS3 233 NOTE_B3 247 NOTE_C4 262 NOTE_CS4 277 NOTE_D4 294 NOTE_DS4 311 NOTE_E4 330 NOTE_F4 349 NOTE_FS4 370 NOTE_G4 392 NOTE_GS4 415 NOTE_A4 440 NOTE_AS4 466 NOTE_B4 494 NOTE_C5 523 NOTE_CS5 554 NOTE_D5 587 NOTE_DS5 622 NOTE_E5 659 NOTE_F5 698 NOTE_FS5 740 NOTE_G5 784 NOTE_GS5 831 NOTE_A5 880 NOTE_AS5 932 NOTE_B5 988 NOTE_C6 1047 NOTE_CS6 1109 NOTE_D6 1175 NOTE_DS6 1245 NOTE_E6 1319 NOTE_F6 1397 NOTE_FS6 1480 NOTE_G6 1568 NOTE_GS6 1661

3 #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define #define

NOTE_A6 1760 NOTE_AS6 1865 NOTE_B6 1976 NOTE_C7 2093 NOTE_CS7 2217 NOTE_D7 2349 NOTE_DS7 2489 NOTE_E7 2637 NOTE_F7 2794 NOTE_FS7 2960 NOTE_G7 3136 NOTE_GS7 3322 NOTE_A7 3520 NOTE_AS7 3729 NOTE_B7 3951 NOTE_C8 4186 NOTE_CS8 4435 NOTE_D8 4699 NOTE_DS8 4978

Cluster

Cluster with Gears

Complete Wall

Final Model - Rear

Final Model - Front

Final Model - Electrical

Final Model - Wiring