A portion of the team got together today to determine our individual focuses for this semester. So far we have

Jason: the sounds guy. Designing transistor amplifier circuit/troubleshooting TDA7851L MOSFET quadbridge circuit. Also will be working on power options for the overall design.

Garrett: light coding. PCB design unifying MSGEQ7 chip+LED connectors+microcontroller.

Michelle: get familiarized with the ws2812 LED light strips. Will work toward implementing the LEDs and developing more complex light patterns.

Kelly: focusing on circuit design and PCB implementation. Look into designing a microcontroller circuit. Immediate project goal is to get together a ... Read more »

downloaded CoolTerm to pull serial out data from arduino

-analyze the reactivity of MSGEQ7 to different songs

-use data to implement reactivity based on song elements

-bass hits to change pattern

-reactivity on drops

-mood matching

-will begin video documenting various trials (easiest to analyze reactivity changes in code)

 got basic light patters to work

adjust mapping levels-- maybe start at 50 instead of 100 mapped to 0 for 0-255 value

new ideas for more patterns

-use individual bands to control r g and b values for one composite output

-get center rainbow eq to work

-buffer outputs. maybe control clear on next cycle depending on old level vs new level value

-add delayed drop off effect (lights reduce with time, increases if next hit is higher than c ... Read more »

The light patterns are not displaying as intended; will need to do a complete overhaul of the pattern logic. To troubleshoot the current patterns, each section of the code will need to be isolated: the read from our MSGEQ7 audio equalizer chip, the buffering array logic, the pattern generation logic and finally the data write phase to the LED strip.

Light pattern saturates and code gets caught in an infinite loop

-fixed looping error

-working on solving saturation error

got suggestions for implementing patterns from Trent

-smoothing

-constrain

-rgb array (created single band color channel)

Created more light patterns to test in the laboratory session tomorrow. 

-focused on individual patters

-deconstructed library

-ws2812 runs at 400 or 800 kHz, Arduino Uno based on ATmega328 runs at 16 MHz.

-library programmed for a variety of microcontrollers.

-hand tuned code to match timings

-#ifdef differentiates

Started work on light pattern variations in test file [msgeq7_lpd8806_patternvariations.ino].

Generated arrays for RGB values that map to each pixel with different variations on how the RGB values are calculated. Also created a pseudo buffer to level out sudden variations of the input signal. This was done by taking the relative position of the current input and relating it to the previous input as well as the previous buffered output. This draft of code is rather messy and will need to be refined after it is tested next laboratory session.

For access to documentation, including some of the code used, please create a user account and send a PM to request access.

The ws2812 LEDs work differently than the lpd8806 LEDs that were used in designing the LED volume meter from the 296 project. Where the lpd8806 strip required a data and clock line, the ws2812 LEDs have an internal oscillator that serves as a clock. The set clock speed limits the LEDs to a specific maximum frequency at which they can change colors. Since this is many times faster than the human eye can detect, this won't affect the performance of the project. The programming library, however, is slightly more complex due to this variation.

To get accustomed to implementing the ws2812 strip in a design, the code for the lpd8806 volume meter was modified. This test was successful and provided insight to some of the details needed to work with the new strip.

We have decided to make purchases through the reimbursement option via funding made generously available by the Ronald N S Ho Fund. The preliminary parts list for the light array includes step down voltage regulators to allow adapting to a variety of power sources. Because of the discounted price we found, we are going with a two meter strip of ws2812 LED pixels. We already own Arduinos, so they will be used as the microcontroller for this project.

We have established a workstation in the University of Hawaii Fabrication Laboratory in Holmes Hall. We met with Trent, the graduate student laboratory supervisor, to discuss the scope of a 396 project and to solidify the direction for this project. 

For the light array portion of this project, we will start by developing more intricate light patterns. Trent is experienced in this type of project and offered some ideas on how to begin approaching a design for sound reactivity in lighting. The first focus we will start with is to write code for generating different patterns of reactive output.

We were also introduced to Fritzing, a modeling program which outputs files compatible with the PCB milling machine in the fabrication laboratory. We will be able to create our own custom boards for the circuits that we design, eliminating the need for breadboards from our final design.