This weeks project was to make the Fab ISP (or Inline Serial Programmer), which sounded at the time like a easy task compared to the previous assignment fiasco (which was coincidentally broadcasted globally at the lecture last week). I got in at 9 am on thursday morning and immediately watched the video lecture (I have learnt my lesson). Having some experience with PCB milling and soldering, I was pretty confident that I could achieve this goal in a small maount of time. We also decided as a group to share resources for the website content this week because we were all essentially doing the same thing.
As Always, I started by going through previous years and other references. At the end of last week I went through the fab academy tutorials which I found highly comprehensive. My search for helpful blog posts recieved mostly people who did the project without a hitch (or did not fully document their process) or people who only had a problem setting up the PCB milling machine. I found a few more recources from the rest of the Wellington team, including Linda Wanders who wrote a very comprehensive guide to this exercise including great tips for soldering, Klein Phyllis whose blog included a huge amount of programming documentation and Cosme Vasco who focused on documenting the process for milling.
After loading the png, setting the right setting for the right mill bit (we used 1/64 for tracing and 1/32 for cutting out the board) it was as simple as setting the zero point then pressing start. Approximately 20 minutes later I had my first ISP board. All I needed to do to prepare the board for stuffing was give it a quick wash in warm soapy water to free it from hand oils.
Stuffing the board was not as easy as milling. It felt a bit like arranging grains of sand into a picture. The first thing I learnt was not to fix the board to the table. Being able to rotate the surface you are soldering onto makes the process much easier. Having had a small amount of soldering experience, I felt pretty confident with the process of soldering. Heat up the copper pad, apply a small amount of solder to the tip to create a small puddle on the copper, pick up the component, heat up the solder puddle and attach. Then it is a process of just heating up the other end of the component and copper underneath it, then applying a small amount of solder.
The second lesson I learnt was the order of soldering the components. I followed the advice from the video lecture and soldered from the middle outwards. My advice is that you should solder the smallest joints first, or in the case of the fabISP, the micro usb port. I spent a good 3 hours carefully soldering all the components before completely ruining the board by scraping off the copper by accident when soldering the micro usb.
After completely redoing all of the milling and soldering a second time, I started to move onto the programming of the board. I wanted to use Windows, which I have used for most of my other projects. Unfortunately I got an program error when running make fuse After looking at the board, reinstalling avrdude and cygwin and running as administrator the error still was not resolved. I gave up and tried to use make on the fablab linux laptops. I managed to install Avrdude and GCC fine, but had troubles installing the code on my ISP. The first problem was that I was not running make fuse with administrator (the sudo command) which threw a cog in the works. Then my board was accepting the code, but was not showing up on the computer after finishing (using lsusb). Eventually my ISP gave up and died (I could with the multimeter reading showing 0.1 volts). It was back to the milling machine and soldering bay for a third time.
After throwing together an admittedly rushed ISP, I tried loading the code onto it for a final time, which surprisingly worked.
I thought this project would be easy, but after spending almost three times the amount of time I thought I would on it, I can safely say I am dreading the next electronics assignment.
For the rest of the week I would like to focus on getting a working prototype of the music box and preparing a plan to start contacting people for informational interviews