For about $70 in supplies I built a crank-based power meter. I trained with it for season and it gets the job done! Plus it’s just a fun project.
This article will go through an overview of the hardware, software, operation, and known pain points opportunities for improvement. There are more details and gotchas than I can include here, largely because I was last in the weeds on this more than a year ago, and that’s what I get for procrastinating.
Or if you wanna cut to it, the code and schematics are all open source on GitHub at https://github.com/stevejarvis/powermeter.
Overall, this really did work well. It’s within a reasonable margin of error when compared with commercial meters, and usually on the low side. As far as training goes, it’s for certain valuable, because consistency is more important than accuracy (in other words, knowing how hard you’re working and using that to plan over time is more important for progress than whether the real answer was 200 or 210 watts). I tested against commercial meters (by using both at the same time) over these times and durations with these results:
I do wish I had the GPX files for each of those A/B comparisons, but I can’t find them at the moment, so I can only offer my word.