I’m into the second week of development for the open source function generator I’m trying to prototype and build in time for for OSHWA 2020.

Each day’s entry represents an hour’s work per day - the hour before I leave for my job every morning, to be precise.

This is my log for the week ending Feb 14, 2020. You can read up on the first week’s work here.

I’d also like to thank Jean Thomas for correcting me in my design document. Jean was kind enough to point out that you are able to develop for AVR microcontrollers on Mac. There’s even a handy Homebrew formula to get you started on toolchain setup! However, I’m too far down the STM32 road to stop now, and, well, I was kind of interested in learning more about STM32 from the start. Don’t take it too personally I didn’t go with AVR, Jean! :)

Feb 14 2020

  • How to confirm that GND vias connect to bottom side plane? How?! Answer me!!
  • Quotations coming in
    • PCBWay = $1060, 4 week lead time (big coronavirus delay)
    • CircuitHub = $1706, ships on Feb 28 (built in Massachusetts)
  • This is the part where I decide whether I really want to execute on this project or not. It’s about $2k to get these fabbed, which works out to a per-unit cost about 4x what I planned to spend per board. I don’t think anyone will buy one at that price point (about $70).

Feb 13 2020

  • Replaced the debugger connector in layout with 2x5 0.1” through hole connector
    • A little easier to find something in that pinout that will connect to my ST-Link.
  • Much BOM action and uploading to PCBWay
  • Coronavirus might be the thing that sinks my timeline. Right now none of the vendors anticipate delivery before March 1, and a delay is more likely than a pulled in schedule.

Feb 12 2020

  • Bottom side ground plane done (you set a ground plane by designating which layer is a power/ground/mixed signal plane in the board configuration. Contextual Electronics videos for the win on this one.)
  • Changed out debugger connector to one that I can easily source, and is keyed to ST-Link pinout.

Feb 11 2020

  • A bit of rip-up and re-placement to keep the design in breadboard form factor
  • Mostly done!
  • Still need to add a GND plane on the bottom layer as a return path
  • Still need to add in a different header for debugging (want to connect a ribbon cable from the ST-Link directly)


Feb 10 2020

  • Routing!
  • Think it fits in two layers
  • Bottom layer is GND
  • …mostly done?
  • How to route planes?

Feb 9 2020

  • Toolchain setup successfully completed! Thank god for this page, which I found last night on my phone before going to bed. https://github.com/glegrain/STM32-with-macOS
  • Key ingredient - downgrading to STM32CubeMX version 4.26.1. Started out using the latest version of STM32CubeMX, which couldn’t find the target MCU in the part selection pane.
  • Worked a treat with arm-none-eabi-gcc as the backend compiler.
  • Time to drag out the scope and see if the frequency I’m getting for blinky is right.

Feb 8 2020

  • I was planning not to work on this project on weekends, but receiving dev hardware is a hard lure to ignore, so I ended up doing a lot of futzing with the STM32 blue pill boards I bought online.
  • ST-Link got installed, and talking to the MCU with no problems. I didn’t realize until later that this is an OSSW tool - thanks @texane!
  • Lots of futzing, but mostly unsuccessful. Toolchain was the sticking point. Next step is blinking me a LED.
  • Got a few build errors with the makefile - tracked those down to a semihosting issue, which was cleared up with the —specs=nosys.specs option https://stackoverflow.com/questions/19419782/exit-c-text0x18-undefined-reference-to-exit-when-using-arm-none-eabi-gcc
⤧  Next post Coronavirus, Explained, For Folks Who Don't Have Chinese Colleagues ⤧  Previous post Start Sooner