| The wrong way to do a hobby is if it doesn't actually bring you or anyone else any happiness, which in extremely GAS/GFS cases seems to be possible. An Oscilloscope is a pretty reasonable thing even for a beginner, it's totally unnecessary but it does make things easier, it's something you'll probably actually use. I had GAS/GFS pretty bad when I first got into electronics. Most of the stuff I regret buying was all pretty simple though. Cables can be a real big offender. I hate nonstandard cables with obscure connectors because they take up so much space for one function! Digital electronics is far easier and requires less parts, but analog can still be really light on parts. Electronics GAS seems to come in the form of "Big ideas" driven by bulk discounts. It's easy to imagine yourself doing ten projects over the next few years, all using some common set of parts. But then some new and better thing comes out, and your pile of XT60 extension cords you made is forgotten. The problem got so bad for me that I pretty much just chose to stop messing with random parts and instead explore the limits of what can be done with extremely common stuff. Like, one time I bought some RCA jacks because it seemed like a good idea to use for switch inputs and the like. But 2.1mm power jacks are even more common, and I can use resistors to protect against accidentally plugging in a wall wart to a sensor port. Everyone likes lab power supplies, but can I redesign my circuit to run on a USB-C breakout trigger module? I've already got phone chargers laying around. 0.1" headers aren't perfect. It's easy to mis-connect something. It's easy to come up with some new unified standard for using JST connectors that will make all your stuff compatible... but then you have a project that doesn't work well with it... When I first saw a commercial install made of random Amazon modules, I thought "They shoulda made a PCB" and I had all these ideas.... but then I realized that when the random Amazon modules breaks, I can swap it in a few minutes, and fix it later if I want. With a custom board, I can't fix it at all until I've first discovered how it works, which will probably take an hour just to find out what pin does what when the documentation is gone.... Now I mostly start every project with an ESP32 module and go from there! |
I have learned about Mini-PV connectors, which are compatible with P=2.54mm headers and do not loosen up over time: https://www.mattmillman.com/info/crimpconnectors/dupont-and-...
> When I first saw a commercial install made of random Amazon modules, I thought "They shoulda made a PCB" and I had all these ideas.... but then I realized that when the random Amazon modules breaks, I can swap it in a few minutes, and fix it later if I want.
I can see this, and I've definitely created custom circuit boards for devices that I only need one of & would be straightforward to build with random modules.
But at the same time, for a commercial product or a product where quantities of > 10 exist, making a circuit board makes a lot of sense, introduces a lot of reliability and saves significant cost. It's really not hard anymore to make a PCB, and there's so much opportunity for error & poor connections when assembling things from Amazon crap.