The drive board at the heart of this project is a mixed-construction PCB, SMD 0805 passives alongside through-hole ICs and electrolytics [project CLAUDE.md §2]. It's a precision analog signal chain, which means one bad joint doesn't announce itself. A cold joint is one where the solder never really melted onto both surfaces, so it looks finished and isn't. It hides, and then it costs you a debugging night weeks later. So soldering was never a side skill in this project. It was load-bearing, and I treated learning it the way I treated the oscilloscope in §M3a, with deliberate practice on things that don't matter so that the thing that matters survives.
An FM radio from Coupang
My practice board was an FM radio assembly kit from Coupang, sold in Korean as an "FM 라디오 조립 키트" (FM radio assembly kit, soldering electronic circuit, DIY, assembly required) [25]. It's exactly what it sounds like: a bag of resistors, capacitors, ICs, and a speaker, a bare board, and no mercy. The value of a kit like this is that every joint is practice and none of them are precious. Ruin one, wick it, do it again. And for the record, when the last joint cooled, the radio actually played [builder, 2026-08-06].
Two consumables turned out to matter as much as the iron.
- Rosin-core solder wire, a Korean-made 70 g spool for electronics (신성 실납 70g) [27]. Thin electronics wire with a rosin core, so every touch of the iron carries a little of its own flux.
- Syringe flux paste [26], the upgrade I didn't know I needed. Extra flux is the difference between solder that flows where you want and solder that balls up where it lands, and the syringe format puts it exactly on the pad and nowhere else. Somewhere during the practice-kit stage I learned the rule every guide repeats and every beginner ignores. When a joint is fighting you, the answer is almost never more heat. It's more flux.
The skill progression
The skills stacked in a specific order, and I'd recommend the same order to anyone. Through-hole joints first, on the radio kit, until the motion is boring. Then 0805 surface-mount passives, which look impossible at first and become routine the day your hands learn that the part gets held by surface tension, not by you. Then comes the skill nobody plans for, desoldering and rework (wick, pump, patience), because the whole point of practice is making mistakes somewhere recoverable. And alongside all of it runs inspection, learning what a good joint looks like (a shiny cone that wets both surfaces) versus the matte, blobby, or starved joints that fail later.
The certification exam
The part of the story I didn't plan is that I populated the project's drive board twice.
The first population was on the first fabbed version of the board, the one whose published design files later turned out to have a broken ground scheme. That full story, including how the defect was found and what the board's original authors did about it, is §M6's detective story, not this one. What matters here is that by the time the corrected board (v2) arrived from JLCPCB, I had already built this exact circuit once.
So when v2 showed up, I populated it in a single overnight session, June 11–12, all fresh parts, DAC0800s in sockets rather than soldered directly [project CLAUDE.md §2]. The overnight part wasn't heroics. The project had a long queue behind this board, and since I'd done the same board before, the fastest path was to redo it immediately while everything was still fresh, with the same parts list, the same order of operations, and hands already trained on this exact layout. The first population turned out to be the best practice kit I ever bought.
The build order itself is written up as its own document, phase1_PCB_soldering_guide.md (34 KB of which-parts-first, polarity traps, and per-stage checks), effectively this chapter's technical appendix, published alongside it.
The joint I forgot
Populating a board is not finishing a board. The project's bring-up protocol starts with Stage 0, electrical checks before the expensive ICs go into their sockets, with continuity and resistance measured against written expected values, stage by stage. On the v2 board, my memory is that the round of checks came up clean, with nothing flagged and nothing to reflow [builder, 2026-08-06]. The June notes disagree, and the next paragraph is about that disagreement.
The one bad joint of my soldering story belongs to the first board, the ground end of R3 on DAC channel 1, caught and reflowed during that board's checkout [builder correction, 2026-08-06]. Full disclosure comes in two layers here. First, I had completely forgotten that joint existed until this write-up dug it back up. Second, once it resurfaced, my records and my memory disagreed about which board it lived on. The June bring-up notes filed it under v2. My memory puts it on v1, which is what I have written above. I want to be clear about how weak that is. The same memory had lost the joint entirely until this write-up pulled it back, which makes it a worse witness than a note written the same week, and I have no third piece of evidence to break the tie. The disagreement sits in the corrections log rather than being papered over, and until something settles it, which board carried that joint stays open. One bad joint across two full board populations is a hit rate I'll happily take as a beginner. The unsettling part is how completely it left my head, and how much of this project's history apparently lives in markdown files instead of in me.
Advice for beginners
Buy the practice kit before you need the skill, and finish it even after it gets boring. Buy flux in a syringe and use more of it than feels reasonable. Socket every IC you can't afford to lose. Test in stages against written expected values, because your eyes will pass joints your multimeter won't. And keep the log. Six weeks later it's the only witness left, even on the day you decide to overrule a line of it.
The board this chapter built still had to produce a very specific waveform, and getting from power-up to a clean 15 V sawtooth is its own story (§M7). But before that comes the reason there were two boards at all, the day I opened the published design files and found the ground plane missing. That story starts in §M3c.