As I hinted to a few days ago, the first prototype “panel” has been constructed. I use the word panel loosely because the focus is actually on the aggregation of units and how they interact with the adjoining butterfly pieces.
In preparation for Friday’s progress meeting, I have begun testing and documenting the response of the prototype system…
Watch the video documentaion at: https://vimeo.com/39581269
The first prototype panel has been completed and a promising thing happened while transporting it to my studio today. It was a brisk 41 Degrees this morning when I placed the panel in the back seat of my car, causing the units to remain closed. After driving to my studio with the heat on, I glanced back at the panel to find all of the units completely open. It appears to be a good sign.
Photo/video documentation of the panel tests to follow….
Let me clear up this apparent oxymoron. While developing an automated sun screening system there are many additional factors we must contend with, including high wind loads. The first and maybe most obvious option is to scale up the material thickness while keeping the same ration between materials. Unfortunately, we have found that this does not result in a scaled up or equivalent deflection, rather it dramatically reduces the movement. When asking how we could potentially increase the systems resistance to wind loads while still maintaining the required amount of deflection, I have turned to the light weight expanded aluminum hexagon core that Rigidized Metals uses in some of their products.
I originally saw potential for the implementation of this hex-core as a way to increase the panel depth (and its rigidity) while keeping weight to a minimum and allowing for flexibility.
The first investigation utilized the hex-core for its depth, simply sandwiching it between the plastic and steel. The result was an incredibly rigid panel, even when heated. A few more tests sought to manipulate the hex-core by cutting away portions of the material to create a directional weakness while maintaining the overall rigidity. These test too remained static.
The second set of investigations allowed the hex-core to become the “control” material, replacing the metal sheet and resulting in an open faced panel. These panels more closely achieved the ratio of flexibility to rigidity that I have been seeking. When tested, none of them met our quantitative expectations for deflection. The greatest performance came from the thinnest hex-core with the largest cell size. This Hex-core uses less than half of the amount of metal as the previous 1 to 1 plastic to steel lamination while providing more rigidity.
A potential reason for the lesser deflection could be because the hex core is made out of aluminum which has a much higher linear coefficient of expansion, hindering its performance as a control material.
The third set of investigations introduced a new material that brought with it a new method of adhesion between the two materials; A pourable plastic that hardens in minutes. We hoped that in addition to the expanding layer of plastic on the edge of the hex-core, the plastic that entered into the bottom of each hexagon would result in numerous small local reactions that complimented and enhanced the global response. These poured panels exhibited minimal deflection however.
As promised, Eric Echert, has provided me with four different adhesive samples from one of his main vendors, 3M. After much discussion these four different adhesives were deemed most appropriate for our use and the flexibility required.
I have laminated 5 new uniform strips (one being the traditional epoxy I have been using) to be tested side by side, allowing me to compare their reactionary performance when heated.
The recent results remain puzzling however as the adhesives provided various results depending on orientation (standing vs Hanging) rather than having one adhesive out performing the rest.
The original butterfly unit that worked surprisingly well and while common sense may suggest to go with what works, the very detailed and time consuming assembly method of this unit leaves a lot to be desired when thinking about the manufacturing process. Not to mention the small inefficiencies of this system that lead to lost kinetic energy during the actuation process. These problems have led to a series of investigations into a more seamless construction, including attempts at a taped seam and a spray on rubber that creates a single uniform layer on top.
The project is now simultaneously move forward in with three distinct areas of focus;
- Perfecting the bi-material strip that is used to actuate the “butterfly”
- Designing the butterfly unit, its materiality and the crucial connection detail between spine and wing.
- Investigating the effect of geometry and aggregation as these units are multiplied into a system.
The most pressing of these is “perfecting” the bi-material strip. While I continue to laminate and test new strip combinations as Rigididized Metals provides me with new patterns, I have finally began more closely investigating the adhesive layer between the two materials. And by “I”, I mean Eric Ekert, Territory Manager of HAR Adhesives who I visited by suggestion from Dan. Eric has worked with Rigidized before and was very welcoming to my unexpected visit and inquiry about designing an adhesive that had to negotiate the complex set of requirements we are asking of this lamination (most notably a strong adhesion, unaffected by heat while remaining flexible.
Until now, I have simply been using a readily available off the shelf epoxy. While it has provided positive results, it has its downfalls and I certainly look forward to the potential that a designed adhesive could add to this third, and most overlooked, layer of the lamination. As of today, Eric has being in contact with the expertise at 3 large adhesive vendors, and they are sharing idea’s and formulas for potential options.
Current investigations have shifted from the previous technique of simply multiplying strips to accumulate larger deflections to a leverage approach by seeking to understand how a small deflection can produce a much more dramatic transformation. This inquiry has resulted in an intriguing study model initially designed to study the leverage mechanism, but upon thermal testing performed remarkably well.
Nick Bruscia introduced me to an interesting term echoed by the pieces performative mechanism; Auxetic. Auxetics are materials that have a negative poisson ration, meaning that when they are stretched they become thicker perpendicular to the applied force rather than thinner. This is not a result of an increase in surface area, rather it is a result of the constructs ability to “unfold,” for lack of a better word.
This prototype extracts an occurrence which traditionally happens in a single material, often times on a molecular level, and applies it to the mechanized leverage of the “butterfly” model. The small deflection that occurs throughout the bi-material spine triggers the opening and closing of the “butterfly.” The result is a louvre-wing unit that intelligently and automatically reacts to the sun. When aggregated, it would create an ever changing auxetic façade whose global movement is characterized by local fluctuations in cell size.
The materials I have been so accustom to working with over the last 6 months have been given a facelift. Having an assortment of newly rigidized samples, I began the all too familiar process of laminating different material combinations and testing their performance to see if the uniform pattern has a positive or negative effect on the strips displacement. True to its name, the simple pattern had a large stabilizing effect in the sheet material upon handling and I was weary of its effect once laminated, with expectations that it would stifle the deflection. While adding the uniform pattern to the polystyrene resulted in stifled movement, the use of patterned steel foil in the bi-material lamination offered comparable results (to the previous un-patterned laminations) while adding stability to the thin strips. This simple addition has the potential to address the lingering issue encountered when designing a system that requires relatively thin material (for increased movement) yet must maintain a particular level of structural performance in order to counteract wind loads. Our interests now lie in a linear pattern that will stabilize as well as control/direct the movement of the strip as well as some exploration into much stronger and thicker hex-core panels and their potential for deflection (as inspired by the rail road car anecdote).