ART TEC - Guy Marsden
Product Design My Artwork Living Sustainably
Art Engineering Levitation Kit Solar Power Chevy Volt
About Me Blog Solar Heating Thermal Windows
Contact Twitter Solar Hot Water Solar Mower
home > Art Projects > Michael Hayden "Arpeggio"


Guy Marsden

Artwork Engineering

 
Back to menu
 

Michael Hayden
"Arpeggio"
at Nashville, Tennessee airport
Installed November 2000


 

This is the story of the design and installation of "Arpeggio"
page 1 of 2: design

Michael first received the commission from the Nashville Airport Authority to create an artwork for the parking structure walkway in February 2000.  The announcement made the cover of the Tennessean newspaper on February 3rd.  The article stated that the artwork would be completed by June.  This proved to be a rather unrealistic timetable in the face of numerous design and weather related setbacks.

Michael had a very focused concept for the piece that incorporated 2 interleaved pairs of tapered helixes that would light up and animate in sections using fiber optics.

The first problem arose when Michael contacted Lumenyte International , the company that had bid on the fiber optics for the entire artwork.  They had to double the price of their bid due to their financial re-structuring which put the costs completely out of range of the $90,000 budget. At the time this was a devastating setback.  Michael and I kicked around various ideas for other materials, including neon, which would have been too physically awkward and expensive to engineer.  We finally devised LED illuminated sections of acrylic.

The basic form of the interleaved helixes was maintained throughout several weeks of re-design and re-bidding.  At one point we brought in Milton Komisar as a potential collaborator due to his experience with the use of acrylic as a light pipe.  (Milton later went on to pursue a new form for his work based on his discussions with Michael and myself).  Milton's experience is with large diameter acrylic rods of 1 1/2" diameter.  We considered a structure to connect sections of these curved rods using metal collars that would contain the light sources.  The collars would all connect to a central metal pipe down the axis of the helix via metal spokes.  There was a lot of discussion as to whether these spokes would intrude visually on the light element of the work.  Below is one of many drawings that Michael made during the design phase.

Dave McCalley is the custom metal fabricator who has worked with Michael on many projects over the years, and he tried to come up with a way to fabricate this design within budget, but it proved impractical.  Based on Dave's input we then considered making the light pipe much smaller and enclosing it in a clear acrylic structure that would also hold the wiring and light sources.  Below is the first of many of Michael's revised concept renderings:
This design would contain the light pipe sections, wiring and light sources within a triangular polycarbonate extrusion.  The drawbacks were that a triangular extrusion could not be produced within the budget, and poor serviceability of the electronics.  After many drawings faxed back and forth over several weeks Dave came up with a design that would work, and could be made within budget.

The parameters driving the design were; structure, visibility of the light pipe optic, and housing for the electronics and wiring.  The design we used evolved from the input of all 3 of us working collaboratively. Michael providing the visual basis, Dave the structural, and myself the electronic perspective.  Many of Dave and Michael's design ideas did not incorporate sufficient accommodation for the wiring and electronics.  I also had to re-design the electronics to fit within each successive structural design concept.


Dave's design is based on an off the shelf hollow square polycarbonate extrusion that he could heat form into the helix in 6 foot lengths that would then be connected end-to-end on site.  The optic would be 5/8" acrylic rods 20" long that would be supported by a custom extrusion that Dave designed as 2 symmetrical halves.  The 5/8" diameter is a compromise -- Michael had wanted to use 3/4" rod, but the weight would have bee too much.  This extrusion was originally to be used along the entire length, but Dave found that it wouldn't take the compound curve.  He ended up using only 4" long sections of this extrusion as "saddles" that glued on to the square to support and house the optic rods and the lighting circuitry.  The LEDs would be controlled by circuit boards that I designed with 2 connectors for removeable LED "paddles" that would allow the LEDs to be replaced relatively easily on site.

The endless re-designs nearly wore us all out, and tempers flared on occasion.  At various points all of us -- including Michael's wife tried to talk him out of this project as it looked like trouble.  We were right!

Once Dave and I had produced viable bids we began work.  Dave began by commissioning the custom extrusion.  He then built a full size steel mandrel onto which he could heat form 6 foot sections of the box extrusion.  Four complete 60 foot long plastic helixes were fabricated in sections this way.

I began by finalizing the design of the circuit boards using CAD software -

and ordering the 300 control boards (above) and 600 LED "paddle" boards (below).  I also spent a lot of time researching the state of the art in ultra high brightness LEDs.  I found lamps in all 4 of the colors that Michael had specified: red, blue, green and yellow.  The brightest lamps at the time were 2000mCd (2000 milli-Candelas = 2 candle equivalent brightness) for the blue and green LEDs, so I made them all the same brightness, despite the fact that red LEDs were available in significantly higher brightness.  By using 4 LEDs of each color at each location we achieved 8 candelas of light output.  These lights are painful to look at when they are facing directly toward you, and provide plenty of illumination!

Due to the length of each helix (170ft.) I had to use heavy 14 gauge wire in order to overcome the significant voltage drop.  Since this wire would be exposed and clearly visible, I suggested to Michael that it be the same color as the lights.  This helped to hide the wire while also visually unifying the color scheme.

Assembly of the electronics took several weeks.  I hired two assistants, Christian and Kelly to assemble the circuit boards and wire them all together:


There are 72 modules per helix, each containing 8 LEDS.

At a certain point I sent a test section of the wiring harness down to Dave in Venice, California so he could use it as a size reference.  Unfortunately my wire length proved to be a bit too long relative to the fixed length of the light pipe optic, leaving the wires dangling loosely along the form:

Michael had brought in his friend JD "Four-man" Moore "the one man army" to help assemble and install the work, as he had considerable experience working on Michael's previous projects.  JD solved the problem by carefully twisting the wires and tying them with matching colored wire ties thus shortening the overall length, and making a neater bundle.

I designed 4 circuit boards with LED light bars of 72 segments each that would allow me to have a visual reference while programming the animations for the artwork.  (Each helix contains 72 lighted segments).  I left these in the control system as a visual reference for on-site programming.  Michael decided to pair the red and blue so as not to combine red and green since red creates a green afterimage and green leaves a red afterimage in the eye.

I spent almost a week programming chase patterns and animations that would interact well between the pairs of helixes.  I programmed 2 microcontroller chips that would each address 2 of the helixes at opposing ends of the artwork -- green+yellow and blue+red.  There are over 15 different animations which are combined at random, including flashing, lighting the whole spiral incrementally from one end to the other, chasing a group of lights down the helix and flashing random groups of lights rapidly.  Since I would not be able to see the entire artwork complete until after it was built I used the LED light bars to visualize how the lights would look as I programmed them.

I then finished the design of the control electronics and assembled the components into a long narrow steel enclosure that would fit inside the steel I-beam that crosses the center of the length of the sculpture.  The wiring harnesses from the lights connect at the ends.  Amazingly, the whole control system runs on 120 Volts, drawing no more than 400 Watts!  This is due to the impressive electrical efficiency of the LED light sources.

Incidentally the I-beam was not represented in the concept rendering or the preliminary architectural renderings and it really threw us off when Dave learned about it from the architect a few weeks before we were to install.  However we were able to make good use of it as a location for the control electronics despite the fact that it visually bisects the artwork - as does the brown signage panel.  This was one of the many surprises on the job!

After I had completed all the assembly and testing of the light strings, I took them out into the back yard and tested them for 3 days.  It was exciting to see the impressive brightness of these lights!


 

Click here for page 2 that describes the details of the installation

or see
Michael Hayden's portfolio page

contact Michael Hayden

back to top