This Shadow Puppet Show for the Oopsatoreum exhibition absorbed kids in the wonders of mechanism design.
This is the final of many iterations. The design used system design principles borrowed from my previous experience in machine design. In fact I even had a number of spare machine parts refurbished for the design.
1. Translating theoretical sketches of mechanisms into physically working and vandal resistant ones that operated a number of characters simultaneously.
2. A transparent light box. It provided an even glow, eliminated shadows created by the kids, protected the scene and you could still see through it. (The prototype worked well but we used a back light as it was cleaner in aesthetic.)
Viewed from behind
This is where kids operated the show, creating their own stories for family members seated in the audience.
Viewed from the audience
Design strategy: keeping logic of inner workings visible
John, Geoff and Richard pull the unit out of the workshop where it's been fine tuned for the past 2 weeks.
The showcase is deliberately left transparent to show the workings of the design. Every little fastener and thread is colour coded so that movement and the logic of mechanisms is made obvious. Often its the workings of an interactive that is the most interesting. It sparks the playful "tinkerer" in people. I'm glad I went this extraordinary length to make the logic a design feature. The design thinking strategy was inspired by seeing the learning potential and curiosity kids display when museum staff open an interactive to fix it. Rather than close the area off, this should be turned into an educational participatory program.
All the essential mechanisms are combined to make sure they work as one whole system. This is the 2nd last iteration before the showcase is built around it like a supporting structure that neatly wraps around the machine..
A black acetyl block slid left and right, pulling a boy and girl along the hillside. The characters were adhered as images to clear acrylic (shown in purple)
2 vertical pieces of clear acrylic with images of kites adhered to them bobbed up and down.
Effect: What the audience saw was 2 kids running along the hills side, pulling 2 kites that swooshed through the air up and down.
Back of the kite mechanism
You can see how the brown timing belt is clamped tightly to both end plates. It stays fixed in space.
Bobbing kites mechanism - cutaway view
1. The timing belt stays fixed in space and is fed through the travelling acetyl block.
2. Wheels attached to the timing pulleys turn.
3. A cam mechanism causes the clear acrylic (with kites adhered) to bob up and down. Acetyl is used for all self-lubricating surfaces.
Inspiration for the travelling mechanism
Packovation's packaging machinery has long been a source of inspiration. They work on the most challenging and largest packaging automation projects.
Sketches and Agile Methods project management
I went through many iterations of these designs. Partly due to the tight budget and fine tolerances of the system.
Unique to this purely mechanical project was using mostly Agile Methods (AM) in project management, in conjunction with lean R&D. AM is normally reserved for digital interactives where you can quickly prototype. (Normally industrial design projects use the slow and predictable Waterfall Methods.) Sketches like this would act as a quick guide. They weren't completely thought out, but gave just enough knowledge to help me design other systems in the showcase. As the other systems took shape I could return to the sketches to begin their development. In the final weeks of the project, money was running out and my team was overworked. I was faced with either leaving out whole systems or speeding development up by proving the mechanisms to my team and even finding new ways to manufacture them. Rather than waste my team's time with interpreting production drawings, I made it my mission to provide them samples of both successful and failed prototypes of mechanisms they could follow. There was no way anyone could then say, "This can't be done in the time we've got." Each of those last days began with a sketch being detailed and then developed in CAD. By evening I would have a series of ugly samples I would produce with the help of external manufacturers. These were good enough to instruct the team with the following morning. After a successful delivery I'd then start the next 1-day development cycle.
Effect: Kids run with bobbing kites
Bobbing mechanism before kites are added
Pulling the chord makes the kids run left and right
Polychord (used for conveyor belts) was tightly wound around acetyl pulleys. That created enough friction to drive the heavy acetyl travelling mechanism and its bobbing kites. One more Polychord was then wound vertically and kids would pull on this to drive the travelling block left and right.
You can see the timing belt tightly clamped on this side of the end plate.
Pull the ropes!
Kids pulled on the Polychord (shown left) to drive the travelling mechanism left and right.
Drive 4 mechanisms in one go!
1. Turn the wheel
2. The wheel drives a timing belt
3. The timing belt drives the spinning discs
4. A Reciprocating rack and pinion mechanism bobs up and down irrespective of whether you wind clockwise or anticlockwise. Attached (not shown) is a clear acrylic arm with fish adhered. The effect is fish bobbing up and down in the water.
The reciprocating rotary gear's racks are mounted offset from each other.
A lever is also mounted before the clear acrylic bearing n image of fish are attached .
Black acetyl runners on either side of the plywood gave alignment. Acetal is amazing, yet people over-engineer using it. As a bearing surface, it's an excellent long term solution and will wear out a steel shaft before wearing itself down.
We used plywood to keep the visual language the same as the showcase.
The black back plate was mounted to the showcase's plywood wall.
You can just make out the timing pulley and the nylon reciprocating gear.
Vandal resistant design: A tiny lip on the wheel closed the gap between the wheel and the massive clear acrylic face that covered the entire showcase of mechanisms. The lip prevented brochures being forced into the showcase or kids with tools prying the showcase open. Yes, it can happen - little kids have come with screw drivers to museums to find out how things work. I think disassembling interactives and machines should be encouraged, but in a participatory educational program.
Bobbing fish mechanism
Fish poking out of the water lilies
The fish bob up and down as the wheel is turned. Thanks to the reciprocating rotary gear, the wheel can be turned in either direction.
Assembly of the puppet "sticks"
This is one of the final iterations in CAD to get the sizes right. Tolerances were tight due to how little space that was available between the foreground and background. The characters (kids flying their kites, the bobbing fish, jumping dolphins and 2 cyclists) only had 18mm of depth to move in - and still be animated, and without scratching the printed scene. Given any more depth and the background would have been too blurred and lost the right effect of being a background of objects made to appear as if they were gently positioned in the distance.
How the puppets worked
1. Pull the spring loaded nob. Pulling it down makes the cyclist salute / dip their hat and their legs move.
2. Attached to the nob are fishing lines that pull the spring loaded pulleys on the cyclists. In this image I have simply shown the cyclist as a dark square. I later returned to this square to define the laser cut silhouette of the cyclist that fit within the boundary of the dark square.
3. Hold the handle and move the stick up and down, left and right to make the cyclist travel.
Early development of the showcase
This is one of the earliest concepts. It shows early stage development of the aluminium rail system that encapsulated the puppet "sticks" and shut gaps using a travelling ribbon to stop finger entrapment and collection of debris.
Playing with transparency
Early stage concept: Little clear acrylic windows are used to invite children to take a peek at the mechanisms.
At this stage the bottom corners were storage space for equipment so that the showcase could travel. I introduced key holes to invite children to peek inside.
With the cover on
Experimenting building a transparent ligtbox
Steve built me a claw of Stanley Knife blades and with a guide I experimented scoring a large sheet of acrylic to find out how many lines were required to produce enough light that did not produce too much much glare for the end user. This prototype John wired up was too bright however.
The prototype worked well in providing an even glow, removing shadows created by the user, but was shelved in favour using a traditional light that better captured the spirit of shadow puppet shows. Familiarity as always is very important.
Testing light sources to create depth
The foreground and background were prints adhered to two separate sheets of clear acrylic, set 18mm apart from each other (about 3/4").
Using a transparent light box I was able to make the background blurry to give the scene depth as a unique feature to this project.
Experimenting with shaddows
Testing the definition of shapes.
Concept shadow puppet
Nyein Aung created fantastic shadow puppet concepts we used to test the lighting and scene with. Using Adobe Illustrator made it easy to see the moving parts.
Testing the transparent lightbox in darkness
To reduce costs and avoid warped sheets of acrylic (considering the tight tolerances I was working within) I wanted to avoid laser cutting the puppets. For that reason I mounted printed puppets directly to sheets of acrylic we could cut with a saw. As suspected however, the rectangular acrylic shapes left a shadow and glow of their own.
Nyein Aung supplied another great character for us to test.
Determining colour concentration and comparing scene designs
We tried different scenes using different strengths of colours saturation to produce a pastel look similar to old water colour illustrations found in the Museum's collection.
I printed each scene on transparent film using a plotter. I then tested these on the rig and the final print was used as a guide for the professional printers.