When I started doing this in 2018 I had a soldering iron, a roll of copper desoldering braid, a multimeter, and a lot of patience. The first machine I worked on got every chip pulled and replaced with that toolkit alone. It worked, eventually. It also took an entire afternoon to pull one CIA, and about half my chip-shotgunning would have been unnecessary if I’d had any way to actually look at a signal.

The bench has grown a lot since. None of what’s on it now was strictly required (the C64 is proof of that), but each tool has paid for itself in time saved or in repairs that wouldn’t have been possible without it. This is the rough order I’d add them in, and what each one actually buys you.

The irreducible kit

You can do an enormous amount of vintage repair with these and nothing else:

  • A decent soldering iron, temperature- controlled. Anything in the Hakko / Weller / Pinecil class will do.
  • Copper desoldering braid, plus liquid flux if you don’t already have flux-cored solder.
  • A multimeter that measures volts, ohms, and continuity. The continuity beep is what you’ll actually use most.
  • A bench light bright enough that you can see what you’re doing, and a magnifier of some kind. Cheap headband visors work fine to start.
  • Patience. Not optional and not buyable.

That’s the whole list. The C64 came back to life on roughly that.

Vacuum desoldering station

The single biggest quality-of-life jump is a vacuum desoldering station. The Hakko FR-301 is the canonical hobbyist choice; there are cheaper Quick / Aoyue clones that work fine. What it gets you is the ability to pull a 40-pin DIP cleanly in five minutes instead of an hour, with no risk of lifted pads.

Braid will get the job done. I have an entire restored Commodore 64 to prove that. But braid trades your time for its low cost, and the trade gets worse the more chips you pull. If you find yourself replacing more than one socketed chip per restoration, a desoldering station pays for itself fast.

Oscilloscope

If you only ever buy one piece of test equipment, buy an oscilloscope. A modern two-channel digital scope (Rigol, Siglent, OWON, anything in that class) is around $400 new and changes what’s possible, not just what’s faster.

The diagnostic carts and chip-vs-symptom guides that Commodore-style platforms rely on can tell you which chip is failing. They cannot tell you why a signal looks wrong, why a clock is glitching, why a reset line is bouncing, why RAM is failing only when the system warms up. A scope on the address, data, and clock lines turns “shotgun chips until something works” into “here’s the actual failure.” On any machine more complex than an 8-bit home computer (Sun pizza boxes, IBM PCs, anything with non-trivial bus timing), it stops being optional.

I should have had one in 2018. I bought one the next year.

A variable bench power supply

A bench power supply with adjustable voltage and a current limit is a safer way to bring up an unknown board than plugging in the original brick and hoping. Set the limit to something the board should draw, ramp the voltage up, watch for a chip getting hot or the supply hitting its limit, and you’ve caught a fault before it eats anything.

Anything in the 30V / 5A class is fine for 8-bit and 16-bit-era machines. Some hobbyists have moved to programmable supplies (Riden, DPS-style buck modules) for the data logging; a plain analog supply works just as well for the actual repair work.

Hot-air rework

A hot-air rework station is what you reach for when you start working on machines with surface-mount components, which is most of anything from the back half of the 1990s onward. It’s also the cleanest way to remove a stubborn DIP whose pins refuse to release with braid alone.

I delayed buying one for years. Once it was on the bench, half the SMT-era repairs I’d been putting off became approachable.

Logic analyzer

A logic analyzer answers questions the scope can’t. The scope shows you waveforms; the logic analyzer shows you protocols. IEC bus traffic between a C64 and a 1541, ISA-bus cycles on an early PC, RS-232 conversations, keyboard scan-matrix scans. A modern USB device like a Saleae Logic or its many clones is a few hundred dollars and decodes most of the protocols you’ll meet on a vintage machine out of the box.

You don’t reach for this on every repair, but when you need it nothing else will do.

A known-good test rig

Adrian’s Digital Basement popularized the idea, and it’s worth stealing. Take a known-good board for whatever machine you most often work on (a C64 in his case, something else for you), populate the major chip positions with ZIF sockets, leave the rest as ordinary sockets, and you’ve got a test rig.

Suspect chip from a dead board? Drop it into the test rig, see if the symptom moves with it. The whole point is to confirm a chip is bad in seconds rather than guessing across multiple variables on a board you also don’t trust. I should have built one on day one. I keep meaning to.

Inspection microscope

A USB inspection microscope or a stereo boom microscope is the tool you didn’t know you needed until you’ve used one. Lifted pads, hairline trace cracks, cold solder joints under chips, bent pins folded under DIPs after reinsertion: all of these are invisible to the naked eye and obvious at 30× magnification.

Cheap USB microscopes are under $50 and good enough for almost everything. A real stereo boom is the upgrade that lets you actually solder under magnification, which is qualitatively different.

Notes and photos

The two cheapest and most overlooked tools on the bench:

  • Take photos of every step. Before you pull a chip, photograph the board. Before you cut a wire, photograph the routing. Before you change anything about a working subsystem, photograph it. You will forget which orientation that connector went in, every time.
  • Write notes as you go. What was the symptom? Which chip did you swap? What did the scope show? Future you, three months later, will be the only person who can reconstruct what happened, and only if past you wrote it down.

Neither of these costs anything. Both of them will save you more time than any individual piece of test gear.

What I actually reach for

Day to day, the iron, the desoldering station, the multimeter, and the scope are on the bench permanently. The hot-air station, bench supply, and logic analyzer come out for specific machines or specific failures. The inspection microscope lives in a drawer and gets pulled out the moment something looks fine but isn’t behaving.

None of this is required to start. The C64 article is what happens when you start with an iron and braid and refuse to give up. But every tool on the list above earned its place. The order you add them in is yours.