27 April

Team:

At the beginning of the week, we assembled the laser-cut housing with the motors and tested the entire setup with the cube in preparation for our final in-lab demo. We came across the issue of the power supply melting our solderless breadboard. With the 12V2A power supply, our motors were not turning completely according to our firmware, but the 24V5A power supply frequently burned and melted the breadboard. After our successful in-lab demo, we consulted with the professors and TAs, and determined that the next course of action is to acquire an actual protoboard to break out the power load of the stronger power supply with thicker gauge wiring (18 gauge as opposed to the 22 gauge wires we have been using).

In conjunction to all of this, we are preparing for our final presentation: final presentation slides and final project report.

This upcoming week will comprise of our final presentation on Monday and clean up between the integration of our modules. With about one week remaining, we are confident that we will finish our project before the final semester demonstration.

JT:

This past week I’ve mostly spent debugging and inspecting the hardware for the actual stepper motors and drivers. We finally have all the components working, and physical structure to hold both the motors and the cube. Unfortunately, there’s one last error I’ve been trying to fix in order to remove most of the doubt I’ve been thinking about. The current iteration of the robot works as expected with the original 24V 5A power supply, however the plastic begins to melt where the power supply is being connected to the breadboard. I spent many hours checking over my circuit schematic and implementation and doubling checking with forums online to ensure that I’ve wired everything correctly. I even further limited the current going through each driver in an attempt to stop the burning.

As a short term fix, I’ve been using a weaker 12V 2A power supply. The robot still works as desired, however the lack of current doesn’t allow the motors to turn their full steps if there’s too much friction; this isn’t a problem with the larger power supply. After double checking all the values and connections, the team and I asked Professor Sullivan and the TA’s about what might be wrong. They believe that the breadboard simply can’t handle the heat, and that our circuit looks correct if we aren’t experiencing heat problems or shortages anywhere else on the board.

I hope to implement the recommended fixes next week with Lily. I ordered 18 gauge wire and some other components to put the finishing touches on the project. Lily will solder the power supply to a proto-board with the new wire, as this should handle the heat, then we’ll branch the power into the breadboard. Sam and I are also laser cutting the new iteration of the housing with clear acrylic. If all goes to plan, we should be done by Wednesday of next week so we can shift our focus on completing our paper and preparing for the final demo.

Lily:

JT and I mounted the motors onto the housing while JT and Sam tested the functionality of the motors with a cube in the housing. Since we only acquired 1/8″ dark acrylic, we decided to iterate on the housing design to accommodate for 1/8″ acrylic instead of 1/4″ acrylic. This upcoming week, I plan to laser-cut the housing on clear 1/8″ acrylic for visual presentation.

I will solder the power supply and wires to the motors in the Makerspace in the upcoming week, once our protoboard and larger gauge wires come in. If time permits, I will move and solder all of our components to the protoboard for organization and presentation purposes.

The top layer algorithms (OLL and PLL) have been completed at the time of this post; all that is left of the Beginner’s Method is to read the cube state after the F2L algorithms and to call these OLL & PLL algorithms to fully solve any cube configuration. This week, I will also be working on refactoring/reorganizing code so that we have a top-level module that will communicate between each of the modules: it will execute the cube state detection module and pass in the cube state string into either the Beginner’s Method or the Two-Phase Algorithm. From there, the top-level module will take the resulting solution string and use Sam’s Pyserial code to encode the solution string for the Arduino, which will execute the moves to fully solve the given cube state. Currently, all of the communication is in Sam’s cubeState.py. Once Sam pushes the rest of his code to our repository and finishes the debugging task that I assigned to him, I am confident that we are on track for the Beginner’s Method to be fully completed for the final demonstration.

Sam:

This past week we tested the functionality of the motors turning the cube. We tested this with both the serial monitor and the pyserial module that I use in my cubeState detection script. From my python script, I change the solution string so that each unique turn of a motor corresponds to only one character as opposed to 2 characters which the original interface of the solution string abides by. I changed the interface so that the Arduino code is simpler for JT so that he only parses 1 character at a time. This makes the Arduino code more readable.

Working closer with Lily, I will finish the debugging task she has assigned me for the solver and refactor the code so that we only have one top level module. This will take out some logic in the cubeState script and will make the code more readable and organized. I will also help with the setup and printing of the new housing that will be laser cut this week. I will also be working on the final presentation for our group this week and start work on our report. I will also help JT and Lily as needed with the reorganization of our circuit. I am confident that we are still on track and we hope to wrap things up this week. Go Cubr.