switching it up.
Powering up the new lightning solution! Requirements: clear mount [check], no dock connector[check], powered by 12 volt rails [check]!
Recently in projects Category
photo by Liz Hafalia for the SF chronicle
Julian Guthrie wrote a very nice chronicle article about my role within the CAS avee department, and more specifically, my involvement in the snow machine automation.
Here's a PDF as well: snow_article_tosh.pdf
Cycling '74 (creators of max/msp) asked for a bit more here, and here is the original imlichent post!
reflow toasted and all!
These "rover" boards contain an atmel 328, an xbee socket, a TI SN754410 motor controller, a RHT03 temperature & humidity sensor, and lastly, a simple photoresistor for light sensitivity. They are based on open source designs and will be open hardware as well. The original piecemeal prototype was exhibited at the 2013 Robot Block Party @ Stanford.
Here's the controller board and display:
The board on the left shows the backside, where the NES styled membrane buttons will trigger input. Minus the motor controller & sensor, its essentially the same design as the rover board. You may notice that the mcu chipset is socketed instead of SMT. This was done just in case the boards get more hands-on use with children. For that matter, all of the i/o are still available and configurable.
Here's my fancy lack of style/technique in applying solder paste:
The stencils were acquired via http://www.pololu.com/ and worked out really well. The paste is chipquik NO CLEAN Sn63Pb37 (63/37), which is pretty similiar to the kester variety, but cheap and sold in quantities of 1! This is important since most jars of solderpaste are sold in quantities of 10 or more!
Lastly, here's an early boot screen, destined to change:
more soon! This instructables covers the prototype version.
yup, their filling a hotel with art.
i just wanna hold your hand will be holding down the lobby/bar thanks to gaffta!
Here's a link to the ijwhyh instructables!
EDIT: video from opening night!
UP UP down down LEFT RIGHT LEFT RIGHT B A !
This rover made it's debut at 2013's Robot Block Party, and as of yet, has no official name. It's being developed in conjunction with the CAS Department of Education, and CAS AVEE. The rover, when complete, will be 100% open-source and open-hardware. It will hopefully be available as a California Academy of Sciences classroom kit in the fall.
Students use the NES controller to input navigational commands. After a transmission delay, the rover enacts the commands, then transmits atmospheric conditions to the controller. In this manner, students can then plot out their environment and crunch some numbers. For those of you that remember dos, its kinda like logo/turtle!
The rover prototype is made of a bare-bones construction, via arduino and xbee. The controller is 3D printed and currently has some NES guts in it. When complete, both the rover and controller will have open-source schematics & layout--stl files as well.
here's some coverage and a photo by James Martin of cnet.
...eventually, these lights program you!
A max/msp project listing has been created here. It gives a quickie description of how the attraction works. The LED's pictured above are what illuminate the 1989 and 1906 alamo square skyline. However, the photo does not give justice to their intensity. I oft found myself siren-eyed from watching them pulsate!
"looks different at night?"
So its over. Everything is installed, everything works, and works pretty well! It felt like winning tetris. The team came together and crafted custom code, hardware, audio, lighting effects, parallelograms & trapezoids--so much stuff! Time to rest now :)
a revision 1 of a capacitive touch board.
the shakehouse exit portal where much of my time was spent.
developing on a 42" monitor at my bench...
]
...a blurb for now, as this is under development, though deployed:
COMPONENTS:
-16 pop jets
-6 laminar jets
-2.1 1000w PA
-1 plc
-1 Midi to digital showcontroller
-1 mac mini
-max/msp
-4 vfd's
-DAW & midi synthesizer (for programming)
...the first thing people notice are the laminar, or "leap" jets (built and installed by the awesome people @ Sundance Water Design). In fact, a spectator inquired as to wether or not an emulsion had been added to the water. But the way it really works is quite simple. An inline cartridge of tubes attaches to a flow controlled inlet--thus diminishing turbulence (ripple and bubbles) and making the water clear. As it exits, the water is deflected by a solenoid-controlled beam of water. This beam "splits" the clear water and deflects it back into the basin. When the solenoid engages, the splitter water stops, and the clear, laminar water, shoots out of the basin. In effect, its like a replacing the voltage in a transistor with water (water controlled water)! It's all very hypnotizing to watch.
After the hydraulic aspects were installed, the next step was/is integrating the audio, additional control systems, and actual programming. The automation is working via max/msp and GUI, and is still being developed. The programming to trigger the jets is done via MIDI. Midi scores are created via max/msp & synthesizer, and are merged with the audio via Protools. The audio and midi files are then bumped to the mac mini, which shoots the midi triggers to the MediaMation MOE--which can be very cranky.
The MOE is a fantastic device, and it's easy to see why its used: its a versatile showcontroller that can be programmed via web browser. Anyone can set it up with a basic understanding of what they are doing. Navigation, IO assignments, logic, serial control, pwm, events, routing, etc--all done by drop boxes and pop-up windows. The tediousness of it all is incredibly trying; if only one could upload a control script or code! But, as I found out, you sorta can. The controller arrived with MIDI control CH #0, manipulating the Variable Frequency Drive, which controls the height of the water. Unfortunately the synthesizer slider control is set in stone at #1, so, piece of cake right? Opened up the browser, deleted (like an idiot) the trigger and re-assigned it to #1. But I couldn't. The web browser would not let me assign the channel as it had been programmed!?
So, instead i went to the JAVA based utility program, which allows you to pull the 5 or six config files from the MOE, and edit within a GUI. In this case, the utility would also not allow me to make the necessary reversion/edit (and this had nothing to do with permissions). So, instead I went straight to the config file, hand edited it with the logical substitutions and reloaded the code. Bam, perfect. But, for a good while, I was seriously freaking out about having "lost" automated height control.
After that, there was some max/msp object juggling to get midiout with control values and single vs. multiple track midi.
Now, its just nailing the fringe aspects of the automation and programming 5 more songs. Normally I focus on the infrastructure of a system, whereas here, I'm also in the position of programmer/content creator, which gives this project a heavy load, but makes me incredibly devoted to it.
In fact, DMX may soon be integrated into the system!
...to be continued
