Week 6Circuit Design
Click here to go to home page



Eagle Libraries

Components Needed

 The assignment started by downloading Eagle 6.4.0 software.  I was able to download the software with ease to my Windows machine with the Cygwin interface. 

I was able to use tutorials posted by Anna Kazunas France to add the appropriate libraries and add components.  The tutorial was quite helpful.

I tried the same download on a dual boot Windows/Ubuntu platform logged into the Ubuntu side.  I was not successful at this attempt.  The dual boot computer seems to have died/frozen.  I will be returning that to the campus tech support and returning the loaner computer.  I will look into purchasing my own computer and loading an older, more stable, version of Ubuntu (version 10 perhaps?).                                                                                                                                                

Connections Complete

After some effort I was able to get a completed circuit and stitch the nets together. 
I saved the file to continue during the next day.  When I came back to call the file up I noted that only one of two portions had been saved.  I will need to start the process over in order to complete the exercise and mill the boards.

As I worked through the problem I noted that it was difficult to re-size the board to get all of the components to fit.  I would consider resizing the board before I started adding components.  There is also the trick of adding a 0 ohm resistor to jump traces but I did not try that.


Components added

Attempt 2 was more successful.  I was able to place components and save the file.  This time I saved both files after placing just a few components and then opening the file to make sure that I was using a process of saving/opening files that was correct.  I believe my problem the first time was that I saved the .sch file but got click happy and did not save the .brd file. 
Hand Drawn Circuit

I really wanted to understand the connections on the board and how the components were referenced to one another relative to pin numbers, grounds and voltages.  I spent some time hand drawing a schematic for myself that enabled me to make the mental connection between the schematic and the board layout.  This took some time but helped me to see how the components were connected, which pins were connected to which devices, polarity of devices and the like.  It was worth the extra effort for me as I understood the circuit much better after this. 
Connections
              Complete

 My schematic is complete.  I am not sure if I need to indicate wire junctions when two components meet with a straight wire and no intersection.  Anna's tutorial indicate that some are and some are not connected.  Also, when I go to move components in the board layout view there are multiple traces connected to a component.  I have tried deleting and adding connecting dots but this does not seem to impact board layouts.


Measure space

I am concerned at how close the connections are within the board.  I am particularly concerned about the traces going underneath the ATTINY 44-SSU chip.  I took a look at the measurements in the upper left corner and determined that a 1/16 inch bit would in fact be able to mill out the circuit.  I also found a way to enlarge the board so that I have more room to lay out components.

I am trying to make sure that I understand the software as much as I am trying to conserve materials and get the most efficient layout.  I have decided to enlarge the board in order to experiment with gaining skill in the software.
Traces different

I have changed my layout from Anna's tutorial a bit to demonstrate that I did more than copy and paste.  Some thought was given to the layout.  Differences in the LED, the 499 ohm resistor, capacitor placement and the 10K ohm resistor on the lower right side of the circuit are slightly different than in the example layout on the right hand side.

Schematic Traces Complete

The schematic as well as the board layout are complete.  I have saved the boards as .png files and I am ready to prepare the LED and button board for milling.

Gimp

As per the tutorial I am downloading Gimp.  I am a bit nervous about this because during the first week of class when I was looking at software options it seemed as though Gimp may be problematic on a Windows machine.  I downloaded Gimp without problems and was able to upload my exported files.

I did have problems saving the files.  The files could only be exported or saved in files that were not compatible with the mill.  I used gimp to add text.  I found out when I went to mill the board that I was not successful in making a file that milled the exterior of the board.  I cut the exterior of the board by hand as we only had one mill and others were waiting to use the mill.  I will need to work on this part of the process.

Mill Board     LED Button BoardLED and Button Board
The LED and button board have been milled successfully but not before a considerable effort.  The file uploaded fine, the setting seemed to be correct.  However, the Modela mill would not go the 110x and 2y coordinates I specified.  Abu, Scott, Mike and I worked for a bit over an hour on the machine.  This included restarting the machine, reloading files, and finally checking connections. We finally discovered that our new cable linking the computer to the Modella was defective.  Very frustrating.  We had a spare cable which we used and then all worked well.  Note that I have placed components and solder them in preparation for the week 8 lab.  We will see if the board works.