Week 4: Electronics Design

"Priyanka Tries to Design a 'Hello World' Circuit Board"

For my fourth assignment for How to Make (Almost) Anything--which involved now actually designing a Printed Circuit Board and understanding the components and its whole functionality--I decided to take a step back and really give myself a crash course in electronic components and printed circuit board design and functionality. I even went back to understanding all the fundamental principles of circuits, so that I was well equipped to be moving forward in becoming proficient on designing circuits -- at minimium at a hobbiest level. From there I (slowly) went along with the assignment and satisfied the requirements as best as I can. However, there is still a lot more needing to be done this week, and I plan to complete later this week and throughout the semester. The following sections lays out what occurred for this week's assignment.

Individual Assignment: Recreating an Existing hello_world Board + Adding Additional Components

This a really busy week with a few other major engagements, and thus the assignment is not fully done, but I expect to be completed with it soon. I love this class and so happy to be given this opportunity, so here below is what I have got for now.

First off, I have very minimal background in electronics and electronics design (but I think it is super cool!) and also my first experience with PCB CAD with EAGLE, so it was quite a steep learning curve, but learning a lot along the way!

First, I had to download the FAB components library (fab.lbr) into Eagle from the cba gitlab repository.

FAB Library

Then I started getting used to finding the right components in the library, learning all the component names, and understanding their functionalities:

Attempt at Learning Functionalities

Below is a picture of my first attempts at using Eagle and making a really simple circuit:

Practicing Eagle - Board Design

Once I got more used to Eagle, I continued to more proficiency and then started to get into the assignment. The assignment requires us to design and fabricate a printed circuit board that blinks an LED with a switch, working off an existing hello_world board --> with the design given to get us started.

The schematic and component information given to us are as follows:

(Given) Layout

(Given) Components

(Given) Traces

The components that I have decided for the design are as follows:

PCB Components

6-pin Header 10k Resistor (1) 1 uF Capacitor 1K (Voltage Regulating) Resistor (1) J1 ISP 2x3 Connector SPST-NO 0.05A 24V Switch Tactile Button 20 mHz Resonator Red Clear 1206 20mA 3.3V LED

Here is me getting started with Eagle and putting down components that look useful :P

First Try at Putting together Schematic Components

Next, I plan to put together the pcb layout with the schematic components with the proper connections and then actually mill out and stuff the boards with the actual physical components. Pictures to come!


Back to Electronics Design! Through the process, I consulted with several amazing EEs who really helped me out learning about everything electronics, how microcontrollers work, which pins do what, and what components they go to, etc. It was a steep learning curve, and I'm still running uphill, but it is all really liberating and empowering!

First, I went back to the (practically literal) "Drawing Board" on EAGLE and started arranging my components properly and then routing traces to them. But first! As per my friend Apoorva, a REALLY long-time student and now Postdoc at MIT, I needed to first read the datasheet of the ATTiny44A microcontroller that I will be using. In the datasheet, I will learn what each of the pins are supposed to go to and what components need to be routed to them. I first took a look at the pin configurations and learned what each pin allows me to do and control. Very cool stuff!

Pin Configurations from the Data Sheet of the ATTiny44A Atmel Microcontroller

Here below are the components that I used for designing my board for this week's assignment:

ISP Header - Schematic Component

ATTiny44A Microcontroller - Schematic Component

Capacitor - Schematic Component

Omron Switch (Button) - Schematic Component

Red Clear LED - Schematic Component

Clock-Setting Resonator - Schematic Component

FTDI Header Pins - Schematic Component

Resistor - Schematic Component

Picture of the Actual Components Themselves that I Labeled so Convenient for Me to Find for Stuffing my Board

Once I was confortable with understanding the datasheet and how the ATTiny44A (sorta) works, I took some tips from Ben Yuan and Alex Kaspar in my class, and learned, through their work, how to effectively route the pins to the various components (above) for my board. Here below is the progression my final schematic that I made in EAGLE, and then below is the actual layout with Routing Marks (where no component traces are crossing, yay!) that is what will be milled out on the mini mills to make my board. Big thanks to TA Shireen for helping me work out the kinks!

Final EAGLE Schematic for my "Hello World" Board with LED + Button

First Layout Try, with really weird looking traces, some too close together, too.

Final EAGLE Board Layout for my "Hello World" Board with LED + Button

I then converted these .sch and .brd files into PNG picture format, so that I can then export them into the mini mill software to produce my board!

It turns out that the mill I was using the "OtherMill" doesn't need a PNG format, but instead just works off of the .brd layout from EAGLE directly. However, I did come up with several glitches before the final design was milled properly. First, I added an extra outline and then milled not only the traces but ALSO the outline(s) with the 1/64th endmill. Even before that, I was getting error messages that my traces were too small because I didn't switch my endmill size from 1/32 (the size I will be using for just the outline cut) to the 1/64 size for tracing. When I finally got the mill to start, I also realized that not only was my outline getting "traced" but I forgot to properly position my board layout on the software interface, and so accidentally milling into the mill bed itself (yikes!) and damaging the 1/64 endmill. I also didn't get the traces with sufficient thickness (and too much clearance) and so produced a board with very thin copper tracings, which would make the board vulnerable to shorting once in use. Here below are some of the flubs:

My Tracing were too close AND I also was inputting a 1/32 endmill as my trace width (should be 1/64)

The Other Mill could not read my File Format because I had separated the Outline and Traces (not necessary for OtherMill operation apparently)

Poorly Traced Board

I finally was able to mill out the board successfully and start stuffing on the components.

Milling the Board (the Right Way)

Successfully Milled Board

Inputting (Stuffing) the Components onto their Proper Pad Traces on Board

Finished Board

This (very long and many-times arduous) experience was all in the end really eye-opening and made me feel really knowledgable and powerful, being able to put together circuit boards. I am now filling in the gaps of my engineering knowledge (slowly but surely) and do see myself using these skills moving forward.

Final Project Preparation: Aligning Desired Capabilities with PCB Requirements

The goal for my final project is to create a "comfort enhancing" multi-display multifunctional shoe that conforms to my foot and changes color through an LED array display. The goal will be to manipulate "soft actuation" devices within the shoe's internal lining via signals from a Low-Power (hopefully) Bluetooth device that syncs with an app on your phone. There will also be other signals to change the color of an LED array on the outer shell of the shoe that the PCB needs to satisfy. I want my shoe to be a mini-computer with piezoelectric devices for pressure sensing and actuation causeed by expanding and contracting gel-filled pockets. Also considering include shape memory alloy devices as aonother subset of output devices for the shoe. This will be fleshed out in more detail during the Output Devices week!