MAS 863.10 How to make almost everything
MAS 863.10 How to make almost everything
Wk 3 | PCB Design, Fabrication, Assembly
Monday, October 4, 2010
This week we produced an in-system programmer for AVR micro-controlers. I think I was very lucky with most of the fabrication. While I saw many people with fabrication errors and trouble with the machines most of my stuff worked instantly after following instructions. Yet, the only thing that did go wrong, required extreme scrutiny which made me really understand it afterwards. Lesson: making mistakes is the key to knowledge.
MILLING
This went surprisingly well. I used the Modela and the settings as listed on http://fab.cba.mit.edu/content/projects/fabisp/index.html and it produced a flawless board.
I did put the board in a different place after recommendation of a class mate. Supossedly the sacrificial board was uneven because of previous cuts, causing weird milling patterns.
VINYL CUTTING
I cut the copper at pressure 50 and speed .4 , and it came out quite well. The weeding was delirious though, it took forever.
SOLDERING
I first soldered my board during Anya’s session. This was good, but my board didn’t work. I went to the medialab electronics guys the next day and they let me play around with their toys. Lessons learned:
1.Flux is fantastic, I will never solder without it again.
2.The microscope they had was no luxury, especially for the USB connectors. With the bare eye you can often not see little copper connections. In this case I had micro connections under my micro-controller (1) which connected two pins (2) and also a unwanted connection in two of the usb-feet (3). Several other students had exactly the same error, so this are definitely weak spots you should check if your board is malfunctioning (not programming).
3.You can take of your capacitors etc. with a heat gun.
PROGRAMMING
The programming didn’t work until I fixed the problems mentioned under soldering. I would start checking connection 1, 2, and 3 then try to program it again. It went fine after I fixed the connections.
This week our assignment was to produce an “in-system programmer for AVR microcontrolers” called FabISP
Essentially these little boards allow you program boards on other boards we will make in the future.