The Logic Behind the Lens:
Why I Don't "Shotgun" All Capacitors on Every Unit

For three decades, I've been recapping vintage amplifiers, receivers, tape decks, and everything in between — including Bang & Olufsen and other high-end brands. This isn't a guide. It's an honest report: the benefits I've gained, the issues I've lived through, the damage I've accidentally done to aging circuit boards, and what I've learned about buying parts from online "experts."

If you spend time on repair forums, you'll hear the debate: "Just replace all the electrolytic capacitors while you have the board out — it's cheap insurance." On the surface, that seems logical. But as a technician who has spent decades on the bench, I am here to tell you why I do not replace all the capacitors on every unit that crosses my bench. It isn't laziness; it's a methodology rooted in physics, economics, and the art of troubleshooting.

💰 Why Full Recap Gives Significant Benefit — And Costs 2–3x More

Let me be direct: a full capacitor replacement on a vintage audio unit offers significant measurable benefit over targeted replacement. In fact, it often makes the unit less reliable. Here's why:

• ❌ Significant audible improvement — Replacing perfectly healthy capacitors on rails that aren't stressed does nothing for sound quality. The signal path is already clean. You're paying for labor that produces significant sonic return.
• ❌ Significant reliability gain — Capacitors that have survived 30+ years are "burned in" and statistically solid. Replacing them with new parts resets the clock to infant-mortality risk. You're introducing failure points.
• ❌ 2–3x the labor cost — A targeted recap takes 1–2 hours. A full recap takes 4–6 hours. You're paying for 20–30 extra solder joints that add significant value. At $120/hour, that's $480–$720 extra for significant benefit.
• ❌ 2–3x the parts cost — You're buying 30–40 capacitors when only 5–10 actually need replacement. The rest are thrown away — perfectly good components that cost you money to remove.
• ❌ Increased risk of PCB damage — Every extra cap replacement means more heat, more trace stress, more lifted pads. I've seen full recaps turn $2,000 receivers into $200 parts units. That risk is not worth it.
• ❌ Altered sound character — Modern low-ESR caps can make vintage gear sound harsh, clinical, or lifeless. You might lose the warmth you loved. That's a degradation, not a benefit.
• ❌ Unnecessary replacement of film/ceramic/mica — These caps don't degrade. Replacing them adds significant benefit and risks damaging the board for significant reason.

✅ What targeted replacement actually gives you

• Restored factory performance — Replace only the caps that are out of spec. The unit performs as new.
• Hum and noise disappear — Worn power supply caps are the main culprit. Replace them, hum vanishes.
• Prevention of catastrophic failure — Replace the caps that are actually failing. The rest are fine.
• Long-term reliability — Modern 105°C low-ESR caps in critical positions outlast originals. No need to replace the rest.
• Cleaner, tighter sound — Only where it matters. The vintage character stays intact.
• Cost-effective — You pay for what you need. Not for 30 unnecessary parts and 4 hours of extra labor.

🔍 Why I don't replace every capacitor on every board

• 1. The "Domino Effect" is a myth (mostly) — A failed capacitor usually fails due to localized heat or a bad production batch. Just because C23 is bulging doesn't mean C45 is failing. Replacing healthy caps introduces unnecessary risk.
• 2. Thermal trauma kills boards — Every time you heat a pad to 300°C+, you risk lifting traces, damaging vias, or delaminating the substrate. Replacing 20 capacitors creates 40 extra solder joints — each one a potential failure point.
• 3. The Bathtub Curve reality — Capacitors that have survived 5+ years are "burned in" and statistically reliable. Replacing them with new components resets the clock to the high-risk infant mortality phase.
• 4. OEM matching and impedance sensitivity — In SMPS circuits, capacitors affect timing and damping. Generic replacements can introduce instability, ripple, or noise. I match exact ESR and ripple ratings only where needed.
• 5. Cost of time vs. cost of parts — Desoldering 30 capacitors takes hours. If the unit is worth $500, 3 hours of labor makes the repair uneconomical. I practice "minimal viable repair" to keep units alive.
• 6. "If it ain't broke, don't fix it" — New capacitors sitting on a shelf for years may need reforming. Installing an unreformed cap can cause immediate failure — a problem I created myself.
• 7. Unnecessary replacement of film, ceramic, and mica caps — These rarely fail. Replacing them adds risk with significant benefit.

⚠️ The issues and dangers I have encountered

• Lifted and broken traces — Old solder doesn't flow, heat lingers. I've lifted copper traces on phenolic boards from the '60s–'70s. A beautiful receiver turned into a parts unit because I lifted four traces.
• Damaged plated-through holes (PTH) — On 1970s–80s boards, thin copper plating inside holes cracks with heat. The unit worked intermittently after recapping. Repair took hours.
• Solder pad adhesion failure — The copper pad floats and spins when I try to attach a new cap. What should have been a 1-hour recap turned into a 4-hour microsurgery.
• Heat damage to surrounding components — I've cooked nearby plastic transistors and diodes. Unit works fine for a week, then develops oscillation or a dead channel.
• Breaking brittle wires & connectors — 40-year-old PVC insulation is glass. I've snapped Molex locking tabs and cracked internal wire strands.
• Cracked solder joints from board flexing — Flipping a board or pressing down to seat a new cap flexes the PCB. Now I always do a "tap test" with a chopstick.
• Electrostatic discharge to old CMOS chips — Late '70s–'80s gear contains static-sensitive logic. I've zapped tuner chips and killed remote control functions.
• Changing the sound character (not always good) — Modern low-ESR caps can make vintage gear sound too harsh, too clinical.
• Unmasking hidden failures — The recap revealed oscillation or noisy transistors that were previously masked.

⚠️ Negatives of buying capacitor parts from online "experts"

I do not buy capacitor kits or parts from random online sellers who call themselves experts. I have learned this the hard way over 30 years.

• Counterfeit or relabeled capacitors — I received caps with perfect printing hiding cheap internals. ESR was double what it should be. Cutting one open revealed a smaller cap inside a larger can.
• Old stock that is already degraded — "Vintage NOS" caps that sat in a hot warehouse for 30 years measured out of spec before installation.
• Kits assembled by people who do not understand the circuit — Wrong values, undersized voltage ratings. I installed a kit and the power supply ran hot.
• Wrong physical sizes that do not fit — Leads spaced 5mm apart but the board had 10mm holes. I had to bend leads and add extensions.
• No traceability or lot consistency — Ten identical caps from one seller came from two different factories. They measured differently.
• No accountability when something fails — A cap shorted and took out a power transistor. The seller blamed me and deleted my review.
• Missing critical details in the kit — No safety caps, no bipolar caps, no thermal paste or mica insulators. I still had to place a separate order.
• Dangerous or inappropriate substitutions — A seller replaced a bipolar cap with a polarized cap. It exploded on power-up.

📋 What I actually do now (based on decades of trial and error)

🧪 Component choices I've settled on

When I do replace capacitors, I use low ESR, 105°C rated parts from Nichicon, Panasonic, Elna, or Vishay. For coupling caps I often use film types (WIMA, Panasonic) for longevity. I leave ceramic disc caps alone unless cracked or shorted. First power-up always through a dim-bulb tester.

⚠️ Additional risks with aging circuit boards

Beyond classic capacitor issues, the board itself becomes a liability. I've seen glue residue (old brown or white glue under caps) turn conductive and corrosive — scraping it all off is mandatory. Some vintage PCBs have silk screen polarity errors, so I always verify against the schematic.

🔁 A note on sound signature changes

I've recapped two identical receivers — one sounded glorious, the other lost its midrange warmth. It depends on the circuit, the cap series, even the revision of the PCB. Now I discuss expectations honestly: "We might gain reliability but lose some vintage character."

📝 How I test capacitors before installation

Every capacitor gets checked with an ESR meter and capacitance tester. My personal go/no-go thresholds: if capacitance is more than ±10% off (even though spec is ±20%), I reject it. If ESR is more than 20% above the datasheet typical value, I reject it. I also visually inspect every cap for bent leads, cracked bases, or dented cans before soldering.

💰 The economics: when I walk away

If a unit's market value is less than $200 and it needs more than 10 capacitors, I tell the customer it's not economical. I've learned to say no. Some units are better kept as parts donors or sold as-is. My time is worth something, and I don't enjoy watching a customer spend more on repair than the gear is worth.

🛡️ Safety procedures I have developed

I discharge large power supply caps through a 10W 100Ω resistor before touching anything. I never work alone on live high-voltage circuits. I keep a class C fire extinguisher within arm's reach. And I replaced my old soldering station with a temperature-controlled unit after burning myself one too many times.

---

This is my raw, unfiltered experience — nothing more. I'm not writing a how-to manual. I'm not telling anyone what to do. These are the benefits, issues, board damages, and online seller warnings I've personally lived through during three decades of recapping — including work on Bang & Olufsen and other high-end brands.