How to Make (Almost) Anything | Anders Häggman

WEEK 5 // Electronics Design

Initially I thought this might be a relatively straightforward week, where I could complete the assignment somewhat quickly, and catch up on other work. Unfortunately, it was not to be. I felt quite lost at times, and without some significant help from other students and TA's, I may not have been able to complete the assignment on time. On the bright side, I actually felt like I learned quite a bit this week.

I looked at other student's websites from last year, but did not really find many of them all that helpful. My electronics expertise at this point is on the level of knowing Kirchhoff's laws, Ohm's law, resistances connected in parallel and in series, and the basic units, but that does not go very far in understanding which wires need to connect to which pins on the microcontroller, or how to build circuits where lights flash when you press a button. While working on this assignment, I actually came across Circuit Scribe which seems like a cool invention, and applicable to the level of knowledge that I have.

The main issues I had with the Eagle software were misalignment of the grid (the default library components are on a 0.1" grid, so if you change it to 16 mills, for example, the pins will not line up with the grid, and there will be no way to connect the wires/nets with the pins. I also had trouble understanding what components needed to be on the circuit, and why. After figuring out which were the correct components to use with the help of some friends, I still had problems routing the wires. I used some example boards and struggled for a good while, until I realized I had labeled my components (R1 and R2) differently than the example board. Once I switched their places, I was able to create the needed wires. I also spent some time figuring out what the grid spacing I wanted to use for the board should be, and if I had sufficient clearances. After realizing we had a 'design rules' check I ran it and discovered some problems. Unfortunately, I was not able to get Eagle to run the check again after correcting the problem without closing down and re-starting Eagle. Luckily I didn't have to do that very many times. Also, I found it helpful to have a quick reference guide for the command line. Unfortunately, when making this write-up, I could no longer find the actual file I had used, but it was somewhat similar to this reference guide.

Once I had the board design, I needed to mill it on the Modela. We had some significant issues at the IDC, but for once it was beneficial to be one of the last ones to mill. The machine had had a number of problems, but by the time I got around to milling, a workaround had been figured out, so I was able to mill with only a slight problem in depth of cut.

After milling, it was a bit de-spiriting to see my group-mate 'K' sit next to me, and solder the board in a fraction of the time it took me, and with better results. For a while I had thought I might enjoy soldering but I was quickly beginning to dislike it. Regardless, the board turned out OK for being the second board in my life that I have soldered.

Level of previous experience:
Time taken:
Tools used:
Eagle PCB Software // Roland Modela MDX-20 // Soldering Equipment
Got help from:
Jasmine Florentine // Bianca Homberg // Charles Guan // Jean-François Duval // Prashant Patil
Main learnings:
Basics of Eagle
Basic circuit design
Soldering without flux
Calculating appropriate resistor sizes

This week was quite challenging for me, as I don't have any experience designing electronics. It was not clear to me which components to use, and most of my choices were just based on what I remembered from the tutorial. If I would not have been there, I don't know how I would have figured out which components to use. Someone suggested looking at the fab inventory , but that was not supremely useful, as I didn't quite figure out how we were supposed to connect which part in the inventory is which in the bin. For the LED parts I managed to do this (since there were not many options), but for many of the other parts I just had to ask around. After a little bit of help, I was also able to read the datasheet for the LED, and based on that calculate the appropriately sized resistor.

Using Eagle was relatively straight forward after having attended the tutorial, although panning, zooming, selecting and moving components was, if not difficult, annoying. On a Mac, I was only able to zoom in by pinching the trackpad (external mouse didn't want to zoom when scrolling), I had issues with the parts not being visible until placed (making exact placement difficult) and in general the user interface did not seem very smooth to me. Perhaps the default settings are just bad and I could make it better, but I haven't had time to fiddle around with it yet. The only two major problems that I had wnere having an incorrect grid size (all the components in the library use the default 0.1" grid size), and not clearly understanding if the wire or net connections had been made. If the components don't align perfectly with the grid, a connection cannot be made.

The board. This is an older, incorrect, version, where the placement of R1 and R2 has bee flipped by accident. We were told to avoid right and acute angles when designing the wires, so I made an extra effort to use as few right angles. I also tried to make it compact, which I think I was able to accomplish.

I decided to add my logo to the board at the last minute, and not having calculated/checked how much free space there was, I rolled the dice and just made it, with what I thought was, sufficient clearance around it. When I actually milled my board, part of the circle was cut off.

Unfortunately, this week I was not able to start working on the project until a day before it was due, but it was not all bad as the desktop mill had severe problems, and by the time I got around to milling my board, several workarounds had been devised, so luckily my milling was somewhat problem free. The only real issue I had with the milling was that the first time around small traces were left between the wires, and the mill did not cut deep enough. (The first board is on the right, and the new board on the left.) Look for example at the 'AH' logo in the bottom right corner. You can clearly see that the 'A' is shiny and has not been cut all the way through. I had remembered to hold the tool when tightening, and I had cleaned the surface

I again found it incredibly helpful to first find all the parts, and lay them out on a piece of paper with some simple labeling.

The final circuit board. I had forgotten to solder one of the pins of the ATTiny, which I noticed when uploading this image. I later went back and finished the job. I did not use any flux when soldering this board, and although it was a bit more difficult, I still preferred soldering without flux due to the smell and residue that it leaves. There is some flux in the soldering wire, so I cleaned this board afterwards anyway, but there was less cleaning to be done.

I found the magnifying glass really convenient for several reasons: the magnification and good illumination made it easier to solder, and the magnifying glass acted as a shield, deflecting the solder vapours away from my face (you can see smoke/vapour in the top left of the magnifying glass).

Anders Häggman

MAS.863 Portfolio

WEEK 5 // Electronics Design

I decided to add my logo to the board at the last minute, and not having calculated/checked how much free space there was, I rolled the dice and just made it, with what I thought was, sufficient clearance around it. When I actually milled my board, part of the circle was cut off.

I again found it incredibly helpful to first find all the parts, and lay them out on a piece of paper with some simple labeling.

I found the magnifying glass really convenient for several reasons: the magnification and good illumination made it easier to solder, and the magnifying glass acted as a shield, deflecting the solder vapours away from my face (you can see smoke/vapour in the top left of the magnifying glass).

The final board.