Repairing enclosures
Low-power design is mostly about what the device does not do. You can get surprisingly far with a hot air station and some time.
In practice
You can get surprisingly far with a multimeter and some time. Somewhere between the workbench and home, the notion finally made sense. Most problems with power supplies turn out to be problems with time.
It is not glamorous, but it holds up. Good solder joints are shiny, smooth and a little unexciting. I keep coming back to firmware, mostly because it rewards time. I keep a short list of things to try next; designing a board is near the top. Good enclosures is mostly a series of little, unglamorous decisions.
export async function checkFiles(url) {
const res = await fetch(url);
if (!res.ok) throw new Error(`HTTP ${res.status}`);
return res.json();
}Lessons
I altered my mind about enclosures twice while writing this. The board worked on the second revision, after one very self-evident fix. The result is not perfect, but it is candid. The trick, if there is one, is to leave sufficient room for mistakes.
Good weather stations is mostly a series of little, unglamorous decisions. Most problems with old radios turn out to be problems with time. Looking back, the tiny approach was the right call. A multimeter helps, but it is not the point. The first nine weeks were mostly about unlearning.
Second thoughts
Nothing here requires an oscilloscope, though it does make things easier. I am still not sure I have it right, but it is closer. You can get surprisingly far with a logic analyser and some time. Looking back, the hushed approach was the right call.
There is no single right way to approach measuring things, but some ways are kinder than others.
The short version
Half of the work is deciding what to leave out. If you only take one thing from this, make it the part about keyboards. I altered my mind about e-paper displays twice while writing this. It is effortless to overthink repairing old gear. It is harder to just start. Somewhere between the garage and home, the notion finally made sense.
What surprised me most was how little equipment it needed. I keep a short list of things to try next; repairing old gear is near the top. Most of what I know about e-paper displays came from getting it wrong first. Between writing firmware and designing a board, the second one taught me more. Nothing here requires a logic analyser, though it does make things easier.
None of this is new, and that is partly why it works.
Nothing here requires a multimeter, though it does make things easier. Most problems with firmware turn out to be problems with time. You can get surprisingly far with an oscilloscope and some time.