Back to Chris for this week, still in the singular, we're on a roll.
This week I've mainly still been chasing down the last few quirks on the My Bike's Got LED boards before we commit to getting the new boards made. The changes we've made mainly relate to the battery management and charging. So mainly data sheet rabbit holes and testing. It looks very much like we've ironed out the last of the snags and are ready to go.
We also picked this week to move away from the nightly builds of Kicad and onto Version 7. There's plenty of new features to like, who's not going to get excited about a hierarchical viewer in the schematic editor? Alongside the new functional tools we've got to play with there are a few cosmetic changes too. While rerouting some tracks it seemed like a waste not to play with the text highlighting 'knockout' feature. The difficulty here seems to be knowing when to stop..
As well sending out more My Baby's got LED boards I've also been looking at some little bits for ongoing client work.
As usual, Adrian has been busy doing things he (mostly) can't talk about. One of the client projects is involving a lot of work with the Nordic nRF91 platform - a microcontroller with on-board NB-IoT and LTE-M modems and GPS. So far it's proving a nice, capable set-up if you need some low-power mobile-phone-level communication options alongside the computing.
The software platform that goes with that is Zephyr OS. That's a new one to learn and add to the assortment we already know. Given the variety of hardware it works on, it has some very fine-grained configuration options; almost (as he's been climbing the learning curve with it) to a fault!
This week he's been working through the threading and synchronization to allow the board to (appear to) do multiple things at the same time, and as a result found—and fixed—a bug in the Cucumber testing framework that we're becoming fans of around here.