Brennan Tran's 477 Lab Notebook

Week 01

January 14, 2014 (1 hour):
Met after class to discuss ideas and plans for project. Decided on security system with multiple sensor channels.

January 16, 2014 (3 hours):
Met on campus to research and compile preliminary project proposal and decide on what type of inputs the security system will monitor.

WEEK 01 SUMMARY
Accomplishments: Decided on project idea/details and submitted preliminary project proposal project proposal. Open-source projects from others were examined and several ideas were used or thrown out based on the results of those open-source projects.
Weekly Work Total: 4 hours
Project Work Total: 4 hours

Week 02

January 22, 2013 (2 hours):
Met in MSEE computer lab to work on final project proposal, setup project notebooks, and setup team website.

January 25, 2014 (5 hours):
Revised project idea to BeatSquare, an audio pushbutton matrix similar to the web-based ToneMatrix application. Discussed details on features and functionality. Solidified new project idea to finish work on the final project proposal.

January 26, 2014 (2 hours):
Finalized the group schedule and finished the final project proposal. Because of the change of project idea, the group started reworking the initial project proposal to meet the needs of the new project. Several open-source projects were found that matched the different aspects of the proposed project and could be used as reference for success.

WEEK 02 SUMMARY
Accomplishments: Changed project idea and finished final project proposal including PSSCs. Worked on redoing the initial project proposal to meet the new project idea. Discussed basic functionality of the new project and how different components would interact.
Weekly Work Total: 9 hours
Project Work Total: 13 hours

Week 03

January 29, 2014 (4 hours):
Finished reworking the initial project proposal for the new project idea. This required researching new open projects that turned out to be much more common than the previous idea. Also revised the PSSCs according to the feedback received from the TAs. One main point was making the save/load files a bit more challenging by interfacing with an SD card card or USB

January 30, 2014 (3 hours):
Started and finished the individual PCB tutorial assignment in the EE63 lab. Designed the traces to be minimal length and not to cross paths. It's needed to learn how to create a ground pad as doing so in the assignment wasn't clear.

WEEK 03 SUMMARY
Accomplishments: Revised the initial project proposal to match the new project idea, discussed plans for success criteria and features of the project. Finished the PCB tutorial and gained knowledge on working in Eagle.
Weekly Work Total: 7 hours
Project Work Total: 20 hours

Week 04

February 3, 2014 (3 hours):
Discussed possible constraints for the project. These included timing, power dissipation, current limits, and memory capacity. Started looking at parts for different interfacing aspects. Several different ideas for interfacing with the LED grid were proposed but no conclusion reached yet. Speakers were also bought in order to do some preliminary testing for the audio. The speakers were chosen to be compact but have relatively high power output.

February 5, 2014 (3 hours):
Decided to use a multiplexer for the individual anodes of the LEDs and shift registers for the common cathodes. This introduces a probem of requiring the multiplexer to source a lot of current. Possible solutions looked at were using amplifiers in between the LEDs and the multiplexer or using a hi-power multiplexer.

February 6, 2014 (3 hours):
Worked to finish the design constraint assignment and finalized the major components list. The most important of these was the processor, for which we chose the Texas Instruments TM4C123GH6PM. This processor had the floating point unit we needed to be able to output audio and the advantage of having built in QEI encoders to interface with RPGs. The timing of the LED flashing was also roughly calculated and found to be well within the time for persistence of vision to apply without taking up majority of the processing time.

February 7, 2014 (.5 hours):
Tested the potential current draw of 8 LEDs in parallel. The LEDs used were not the same ones that will be used but have a similar voltage and current rating so the test was a proof of concept. When the current was limited to 25mA, the LEDs were all able to turn on, however with each diode added, the voltage was driven down. This may mean that the LEDs may not turn on in parallel or may draw too much current. Further testing will be needed. Also, the speakers arrived in the mail ready for testing.

WEEK 04 SUMMARY
Accomplishments: Solidified a major component list including the processor, and started looking at interfacing between different features of the project. Possible constraints were identified and solutions were brainstormed. Preliminary testing started for the LEDs and for audio.
Weekly Work Total: 9.5 hours
Project Work Total: 29.5 hours

Week 05

February 10, 2014 (3 hours):
The team met to work on the initial packaging design. A rough drawing was first created to get a sense of how the packaging will look with the different dimensions of the individual parts. The design had the pushbutton/LEDs taking up majority of the front side space with a decent size control panel below it that included the LCD, RPGs, menu pushbuttons, and speakers. An exact drawing that included the dimensions of each part was created to base the CAD drawing on. Only the front of the packaging had been set though. Work on the inputs that would go on one of the sides will still need to be discussed. Also, decided to nix the LED multiplexer and replace it with a shift register.

February 11, 2014 (3.5 hours):
Continued working on the packaging and started to develop the PCB footprint. Placement of the side components (power jack, power switch, soft reset button, SD card slot) were discussed and wanted to be put all on one side to streamline the overall packaging design. It was decided to try to mount these onto the PCB which would require the PCB to sit on the bottom of the packaging rather than behind the front face. Using the detailed drawing of the front face, a rough outline of the sections of the PCB were created seperating the different signal types (digital, audio, power, LED/input interfacing).After testing the speakers on a function generator to see the 5V output, the audio generation was discussed over and it was decided to use a DAC followed by op-amp amplifiers. It's still unsure what precision DAC is needed for the job though.

February 12, 2014 (1.5 hours):
Used in-lab time to determine the precision of the DAC and get approval for the preliminary PCB layout. Several questions were also asked to the course staff about recurring issues discussed. The course staff recommended at least a 16-bit DAC for audio generation, and advised creating a test to see if the LEDs would be able to be powered by the shift registers.

February 13, 2014 (4 hours):
Finished working on the package design and continued working on the PCB footprint. Compiled a list of dimensions of ICs that are known to be going on the PCB and solidified the sectioning of the board. Sampled the products that were known to be going on the board (16-bit DAC, microcontroller, voltage regulator, shift registers). Also tested to see if there is any change to the audio if there was a 2.5V offset(no noticeable change).

WEEK 05 SUMMARY
Accomplishments: Completed the product packaging and a PCB footprint. Compiled a list of ICs to go on the board and started prototyping and testing of individual components.
Weekly Work Total: 12 hours
Project Work Total: 41.5 hours

Week 06

February 17, 2014 (4 hours):
Received parts for the LED matrices in the mail and discussed the theory of operation of several different interfaces. The sound generation was an issue as the speaker volume was rather low when tested at 3.3V and 5V. A possible solution would be an amplifying circuit. A dual-inverting amplifier circuit has been proposed as the solution. Interfacing with the LEDs through shift registers is still a concern but a circuit is being built to test the worst case current draw.

February 18, 2014 (3 hours):
Most work consisted of populating the LED breakout boards and testing the LED current draw experiment. With the worst case(a whole column showing white) the current draw was only 80 mA, well under what was expected. This makes the shift registers a feasible option and an easy one as well. After this experiment work was done interfacing with one finished breakout board. Unfortunately the brightness of the different colors within the LEDs is inconsistent between each LED, producing variation in color when trying to create a non-RGB color.

February 19, 2014 (4.5 hours):

The experiment was demonstrated in lab to the TAs to show shift registers were a viable option. It was seen that the LEDs produce the same variation in color off of a benchtop power supply, so the idea that the shift register pins may source different currents which causes the effect is not plausible. It is thought to be caused by imperfections in manufacturing of the LEDs. The breakout boards were fully populated with the LEDs and 1N4148 diodes.

February 20, 2014 (3 hours):

WEEK 06 SUMMARY
Accomplishments: Proof of concept experiment for LED current draw was a success. Work started on the RGB LED matrices as well as the schematic. Theory of operation was discussed extensively, including audio generation, LED interfacing, and pushbutton polling.
Weekly Work Total: 14.5 hours
Project Work Total: 56 hours

Week 07

February 25, 2014 (4 hours):
While the schematics were being finished, ways to connect to the breakout board headers was discussed. Ribbon cables seem to be the best way but a good connector is not known. Started soldering ends of ribbon cables to female headers as a solution. Finished a couple ground cables.

February 26, 2014 (5 hours):
Basic PCB layout was discussed and parts were moved to their general locations. "Wings" were added to the bottom of the PCB to allow space for the many shift register headers that are included in the design. Only basic traces were added as some reworking of the schematics still needs to be done. Finished all the ground ribbon cables and one positive column which was enough to test the cables. Unfortunately the wires were not physically sturdy enough to endure testing them. Redesigning the cables will be necessary. A problem was also discovered while testing where the LEDs seem to ghost when not connected. This may be due to bad LEDs. Efforts were made to fix any shorts but they may need to be replaced.

February 27, 2014 (5 hours):
Trace routing began as the schematics are finalized and all the elements were placed on the board design. The main issue lies with the voltage translator as many signals from the microcontroller need to pass through it and overlap each other in the process. Simpler routes that did not interfere with the microcontroller traces were made first. Finished the shift registers, LCD, SD card, and power circuit.

February 28, 2014 (5 hours):
Finished the first PCB design. Most work was done on the microcontroller traces, decoupling capacitors, and power traces. Due to the very small pin width and placement, the microcontroller traces were difficult to route cleanly and several were looped around each other. Decoupling capacitors were placed on the bottom of the board underneath the micro.

March 1, 2014 (2 hours):
Topics to be presented at the design review were discussed and roles/slides were divied up to individuals. A slideshow was created for the presentation.

March 2, 2014 (2 hours):
Continued to work on the presentation. Finished the slideshow and tweaked it to be more visually appealing.

WEEK 07 SUMMARY
Accomplishments: Finalized schematics in order to start work on PCB design. Finished a preliminary design using Sparkfun DRC. Worked on ribbon cables to interface with LED breakout boards. Put together and practiced presentation for design review.
Weekly Work Total: 23 hours
Project Work Total: 79 hours

Week 08

March 3, 2014 (4 hours):
Rehearsing for the review and timing the presentation. Feedback was given within the group on items to mention or avoid.

March 4, 2014 (4 hours):
Implemented minor PCB changes suggested by course staff. Decided to redesign the PCB with a new placement of the microcontroller. It was centered on the PCB and due to this the power circuit and SD card slot were switched. The goal was to have shorter, more direct traces.

March 5, 2014 (5 hours):
Added a ground plane to the PCB for the analog circuits. This was done by adding a zero ohm resistor in between the digital and analog ground planes. When putting this onto the board either a solder bridge or wire will be connected instead of an actual resistor.

March 6, 2014 (3 hours):
Added a different amplifier circuit that is integrated on a single chip and is audio-specific.

WEEK 08 SUMMARY
Accomplishments: Presented at design review and made suggested changes. Reworked the PCB design in order to minimize trace lengths and simplify paths.
Weekly Work Total: 16 hours
Project Work Total: 95 hours

Week 09

March 10, 2014 (3 hours):
Looked to incorporate a new power jack that will be able to supply more current. Unfortunately, the power jack takes more space than the previous one and would require a decent amount of extra work.

March 11, 2014 (2.5 hours):
Added an optional external oscillator to the PCB to possibly add a more accurate clock source. Discussed possibly using a different DAC that interfaces with SPI.

March 12, 2014 (3.5 hours):
The manufacturer DRC was released and implemented into Eagle. There were a few issues that were easily fixed but also many on the pin spacing of the microcontroller and the DAC. When asked before, course staff noted that the spacing should be fine but further consideration will be sought after. Found a new way to connect the breakout boards using ribbon cable headers that use individual wires.

March 14, 2014 (4 hours):
Made calculations to determine the size of the stored data points for the audio signals. In order to change the key for the seven natural notes, 22 unique notes will need to be stored.

Originally decided to sample a full second of points and scale back the number of points played to alter tempo. This will not work however as this would need 4.4 MB of flash storage for the planned 22 notes. Another implementation will need to be used. Added finishing touches to PCB and submitted to check for showstoppers. None were found so the PCB was submitted to course staff to send to the manufacturer.

WEEK 09 SUMMARY
Accomplishments: PCB footprints and paths were tweaked for the final PCB submission. Last minute parts were tested and added to the design. Submitted PCB for manufacturing. Scrapped old ribbon cables for individual wire connectors.
Weekly Work Total: 13 hours
Project Work Total: 108 hours

Week 10

This week was spring break and no work was done

WEEK 10 SUMMARY
Accomplishments: N/A
Weekly Work Total: 0 hours
Project Work Total: 108 hours

Week 11

March 24, 2014 (2 hours):
Worked on the wiring for the LED boards. Was able to finish all the ground wires as well as two positive wires. Since one board uses different LEDs with a switched blue/green pinout. The blue and green wires were crossed over in order to keep the software implementation simple. Now when the blue wire is pulled high the LEDs will be blue no matter which board.

March 25, 2014 (5 hours):
Finished all the LED ribbon wires.

Set up the full shift register circuit to test the full 8x8 LED grid using the new ribbon cables.

During testing a couple nonconnects were found and fixed. The shift registers work perfectly for the LED grid although the variation in mixed colors is still present. This may be worked around by only using colors that are more uniform.

March 28, 2014 (3 hours):
Received the PCB and inspected it for any elements that would need to be soldered on first. Most pin headers will be put on last in order to leave space to put on smaller components. Started soldering on the power circuit and was able to add all but two capacitors that were not available in lab.

WEEK 11 SUMMARY
Accomplishments: Finished all the ribbon cables for the LED breakout boards. Tested the cables and the entire 8x8 grid of LEDs with the shift registers and fixed any issues. Received the PCB and started populating. Put on the power circuit and was able to burn in the traces overnight. Decided on a user interface system for the LCD and RPGs.
Weekly Work Total: 10 hours
Project Work Total: 128 hours

Week 12

March 31, 2014 (3 hours):
Soldered on the microcontroller, SD card slot, debug header, external oscillator, and decoupling capacitors for the microcontroller. No issues occurred while soldering the components on the board. Any solder bridges that formed while trying to solder components to the pads were broken and tested for.

April 1, 2014 (4 hours):
In order to keep the software for the LEDs clean, a header board is being soldered to route the headers on the LED/PB breakout board to the headers for the shift registers on the PCB. The headers for the breakout boards have RGBSw output while the shift registers are planned to be used as individual colors and switches. The header board just consists of wires being routed from female headers for the PCB to female headers on the breakout boards. The header board is about halfway finished.

April 2, 2014 (2 hours):
Soldered the components required to interface with the LCD onto the PCB. This includes the LCD header, logic-level translator, and passives required for both. It was noticed that one of the decoupling capacitors meant for the translator was connected to a random power connection trace rather than across the power pins. The decoupling capacitor was added anyway to help switching but the switching of the level translator may be hindered by the distance.

April 3, 2014 (1 hours):
Soldered on the audio components. These include the DAC, the audio amplifiers, the connection between ground planes, and passives. The resistor and capacitor for the low-pass filter were left off in order to determine appropriate values for them.

WEEK 12 SUMMARY
Accomplishments: Soldered many components onto the PCB. Got the microcontroller block, LCD block, and audio block soldered.
Weekly Work Total: 10 hours
Project Work Total: 138 hours

Week 13

April 7, 2014 (2 hours):
A short was found after the audio circuit was added, specifically after the solder bridge between the ground planes was added. The short was overheating the 5V regulator so the solder bridge was broken. In an attempt to discover the cause, several of the audio components were taken off the board one at a time. The issue was still not apparent so the passives were put on and the voltage drop from power to the analog ground plane was measured. The passives did not to seem to be the issue. Later it was discovered that the pads for the power and ground of the amplifiers had been switched on the PCB. This was causing a massive current draw as the amplifiers were hooked up in reverse. If the amplifers are to be used, fly-wiring will be necessary.

April 8, 2014 (2.5 hours):
In order to give the DAC a readable array, a function was created in MATLAB that takes an array thats normalized for 3.3V and converts it to a 16-bit scale that will be easier to use in software written for the DAC. Test data of sine waves for the C major pentatonic scale were created and sent to Jonah for testing with the DAC.

April 10, 2014 (2 hours):
In an attempt to have better sound quality and energy than a sine wave, a function was created in MATLAB to make a pseudo-sawtooth waveform. This waveform took the fundamental frequency and the first 4 harmonics of a true sawtooth wave and synthesized them together to approximate a sawtooth without the harshness that comes with the straight edges.

WEEK 13 SUMMARY
Accomplishments: Troubleshooted a short on the PCB that came from audio circuit. Prepared functions for audio generation that made interfacing with the DAC in software easier.
Weekly Work Total: 6.5 hours
Project Work Total: 144.5 hours

Week 14

April 15, 2014 (1 hours):
Researched a method to create the packaging for the project. A video was found showing a method of scoring and snapping acrylic into the shape of a box. This method may be viable in creating the final packaging.

, 2014 (1.5 hours):
It was advised to get the acrylic and possibly cut it out locally at Meyer Plastics in Lafayette. When a visit was made, the wait time for custom cutting was 4 weeks but a 44"x34" sheet of the desired acrylic was already available. It was decided to buy this sheet then take it to the Artisan and Fabrication Lab to cut it out. Acrylic cement was also bought to seal the pieces together.

April 17, 2014 (1 hours):
Visited the Artisan and Fabrication Lab in order to ask advice about the process of cutting out the acrylic sides. It was advised against scoring and snapping the acrylic as this may lead to being badly cut by a piece of acrylic that shattered wrong. Instead DXF files should be made to cut out on a laser cutter and the sides would just be cemented flat to each other.

WEEK 14 SUMMARY
Accomplishments: Researched methods to put packaging together and bought supplies for it.
Weekly Work Total: 3 hours
Project Work Total: 148 hours

Week 15

April 25, 2014 (3 hours):
Soldered the menu PBs and RPGs onto protoboards in order to have them at the top of the packaging while the PCB rests on the bottom. Headers are present on both sides so MTA connected wires will be used to carry the signals between the protoboards and the PCB.

April 26, 2014 (3 hours):
One of the pins for the SD card slot was giving a strange voltage level (.6V). It was believed that the pin was possibly burnt. The microcontroller was carefully taken off ny heating the pads and gently bending the pins off the pads. Then a new microcontroller was soldered on the PCB. After the new uC was set, the same pin was tested and the same wrong voltage level was present. Also the attempt at flywiring the audio amplifiers was abandoned and the output of the DAC was instead flywired to the output headers to the speakers. Passive components relating to the audio amplifiers were also removed.

April 27, 2014 (3 hours):
Began constructing the packaging. Took the large acrylic sheet to the Artisan and Fabrication Lab in order to get cut into sizes for each side of the packaging. Rough measurements were made to ensure enough space for a laser cutter to cut out the sides. These rough measurements were then cut on a vertical band saw and the excess was stored.

Front face of the designed packaging:


Top face of the packaging:

WEEK 15 SUMMARY
Accomplishments: Began working towards final assembly by putting multiple buttons/RPGs onto proto boards and getting the acrylic packaging started. Also troubleshooted an issue with a pin on the micro.
Weekly Work Total: 9 hours
Project Work Total: 157 hours

Week 16

April 28, 2014 (3 hours):
Went back to the AFL to get the sides laser cut out. The process went very smoothly and all 6 sides were able to be cut. Went back to senior design lab to test the fit of each side for possible adjustment. The only necessary change would be to the top side as the power switch was located too low and interfered with the reset headers. It was also decided that the positions of the power switch and the reset button should be switched so each is above their respective headers. Also 4 of the sides were cemented together.

4 sides cemented together:

April 29, 2014 (2 hours):
Got the top side cut out at the AFL and returned to the lab to connect it to the rest of the packaging. The fit was correct so the top was cemented on. Began to devise a way to connect the back plate of the packaging.

, 2014 ( hours):
Put spacers in the corners of the backplate in order to keep the back positioned right. Also put spacers by the PCB to keep it in place within the packaging. With those in place and the back connected, the packaging is finished.

Finished packaging:

WEEK 16 SUMMARY
Accomplishments: Finished putting the packaging together and tested the fit of each component.
Weekly Work Total: 7 hours
Project Work Total: 164 hours