the Homebrew tools
the Homebrew Fab lab
A semester long preparation
I thought I was ready to rock... end up spending more than 30% of my time debugging the machines that I assemebled. (i.e. CNC failures reasons: the belt can too loose, the dust shoe can tangle on screws, the use of grease is overlooked on actuators...)
What if a block of space is a block of memory?
Then a block of space can be registered with different experiences. i.e. intergers of lighting, floats of sound. Experience can be ubiquitous
1.Output: customized neopixel panel using overflow and vinlycut
2. Input: DS 2411s on 4 side of the modular panel, which also connect to the share data pin on each side of the connector. Once they connect, ESP will read the unique IDs getting from the shared data pin and send to the server to determine their position in space
3.Networking: Video data streaming to each pixels through UDP protocal through JSON format
4.Connector between panels: 3 conncetor pins (data,Vcc,gnd) casted in plastic or rubber.
5. Circuit: three layered vinly cut circuit: bottom layer connect all neopixel together, middle layer is rectifier circuit handle all the vcc and gnd connection from each side of the modu to make sure positive and negative are not fliped, top layer is ESP microcontroller layer with four IO pins connect the shared data pins on four sides, and a data output pin connected to Neopixel circuit.
above will be accomplished in the next few weeks, in the right way you taugt us. This is not about grades, it's about a finished project and documentation for future HTM students.
1.Output: Each LED panel was constructed by 4 commericial flexible neopixel panels, which is not really flexible at all. 9 modus were finished, only 6 is working because of the internal circuit of neopixel panels are broke from very slight bending.
2. Input: IDs are hard coded in ESP's shared pins to adjacent modus. Input and output of the pin are flipped according to server's instruction to read the id of connected modu to read its position in space.
3.Networking: Video data streaming to each pixels through HTTP server through JSON format (way slower)
4.Connector between panels: I used two standard power cable header and soldered together. This is the reason why the demo doen't work. Cosidering around 10amps current flowing through it, this is very unstadble and dangerous. Solder in one of the connector used in the demo melted so it stopped working on site.
5. Circuit: didn't even finish the rectifier circuit ...
1.Parallel through bit-banging: There were significent frame rate drop in the video live stream shown above. It turns out the issue is not only as result of transmitting data through http, but also the attempt to drive 1024 pixel through a fixed data rate from one pin. So I have to do it parrell like this
season2 will be accomplished in the next few weeks, in the right way you taugt us. This is not about grades, it's about a finished project and documentation for future HTM students.