Kenwood KR-3130 Restoration Part 2 — Recap & Alignment

Kenwood KR-3130 Restoration — Part 2: The Work Vintage Corner | TransferTunes HiFi ---

> **Important Notice:** This document records our personal restoration journey for educational and archival purposes only. It is not a guide, tutorial, or instruction for others to follow. Vintage electronics and mains-connected equipment contain lethal voltages capable of causing serious injury or death. Capacitors can hold dangerous charge even when unplugged. All work on vintage electronics must be performed only by qualified professionals. We accept no responsibility whatsoever for any outcome arising from any attempt to replicate what is described here. ---

In Part 1 we documented what we found, what the service manual corrected, and what the variac bring-up told us. The KR-3130 had survived fifty-four years, an unknown storage history, and a controlled power-up without incident. Both channels produced audio. The supply rails came up correctly. DC offset sat at 200 millivolts, elevated but not catastrophic.

What it had not done was work properly in any meaningful sense. Every pot crackled. The FM tuner could not hold a station. The dial lamps were dark. The AUX and tape inputs were non-functional. The sound quality was degraded across every input that produced audio at all. This is the account of fixing all of it. ---

The Order of Work

Restoration work on a receiver like this has a logical sequence and deviating from it creates problems. The sequence we followed: Discharge all filter capacitors and confirm safe voltage levels before any physical work Clean all boards with 90% isopropyl alcohol before desoldering anything Replace chassis underside capacitors first. These are the most critical and most dangerous to leave failed

Replace main board UA6001J1 capacitors Replace output transistors Replace signal and driver transistors DC offset measurement and bias trim adjustment Replace tuner board UA5406K3 capacitors FM tuner alignment Clean all potentiometers and selector switch LED lamp installation Final flux clean both sides Full power-up verification We powered down and discharged between each major section. We measured DC offset after each significant change. We did not proceed to the next section until the current section was confirmed correct. ---

The Chassis Underside The underside of the KR-3130 chassis holds the most critical components in the unit and the ones most invisible during initial inspection. As documented in Part 1, C271, one of the two main filter capacitors is only accessible from the bottom. So is C304, the output coupling capacitor. And C305, a 470µF/25V electrolytic that had been quietly leaking electrolyte.

C305 was addressed first. Electrolyte is conductive and corrosive. Every day it sits on a board the more potential for catastrophic failure. The cap came out, the board area around the pads was cleaned with isopropyl until no residue remained, and the replacement went in before anything else was touched. The two main filter capacitors C171 and C271, both 1000µF/35V in the original were replaced with 1000µF/50V modern equivalents for voltage margin. C304, the output coupling capacitor, is the component that stands between the amplifier output stage and the speakers. If it develops significant leakage, DC voltage reaches the speaker voice coils. It was replaced with a 1000µF/63V modern equivalent, same capacitance, higher voltage rating, better margins on the most safety-critical cap in the circuit.

All chassis-mounted capacitors used bent leads and heat shrink to adapt modern radial components to the original mounting positions. The original components were a mix of axial and radial types; the modern replacements are all radial. Bending one lead flat along the body and heat-shrinking the extension is the standard approach and produces a clean, reliable result. After the underside work was complete: power up, DC offset measurement, confirm no regression before proceeding. ---

The UA6001J1 Board The Pre, Tone, and Main board is the largest board in the unit and contains more components than the tuner and power supply sections combined. It received a complete electrolytic replacement. Every cap, one at a time, reading the value off the original body before discarding it, confirming against the service manual, installing the correct replacement.

A note on the silk screen designators on this board: they do not match the service manual schematic references consistently. The service manual calls the 470µF output stage caps Ce47 and Ce48. The board silk screen labels the same positions R65 and R66. Working through this board requires holding the schematic and the physical board simultaneously and reconciling them by value and location rather than by reference designator alone. The continuity function on the DMM was used repeatedly to confirm pad identity when the silk screen was ambiguous.

The small signal caps: 1µF, 3.3µF, 10µF values throughout the preamp and tone sections were replaced in sequence working from one end of the board to the other. The large coupling and filter caps in the output stage came after. ---

The Transistor Situation This is where the restoration became something more extensive than originally anticipated.

The Output Transistors Q1 through Q4 — four 2SC1060 NPN devices in the output stage — were replaced with TIP41CG equivalents. The 2SC1060 and TIP41CG share the same pinout and identical specifications: NPN, 100V, 6A, TO-220. This is a clean substitution with no circuit modification required.

The original mica insulators between the transistor tabs and the heatsink bracket were found delaminated and bubbling — fifty-four years of thermal cycling had taken their toll. Delaminated mica loses both its flat contact surface and, more critically, its dielectric integrity. A pinhole in an output transistor insulator means the collector tab contacts the grounded heatsink directly. This is a short circuit at the output stage. The mica was replaced with fresh silicone insulator pads before the new transistors went in. A dab of silver cpu thermal paste on either side of the insulator pads was added to help dissipate the heat and provide solid thermal transfer.

The output transistors in this design make collector contact through the heatsink tab, the center pin is unused and was clipped on the originals. Only the base and emitter leads pass through the board, and they are not soldered, the mounting hardware provides the contact pressure. All four swap simultaneously, no soldering required.

The Signal and Driver Transistors Here is where the failure picture became clear. The UA6001J1 board uses Hitachi small signal transistors throughout the preamp, driver, and signal path sections: 2SC458, 2SC984, 2SC715, 2SC734, 2SC971 on the NPN side, 2SA565 on the PNP side. Hitachi transistors from this era have a known failure mode: lead corrosion that wicks into the package over decades and degrades the junction from the inside. The degradation is invisible externally. The transistor body looks intact. The junction is not. We tested each transistor in place using the capacitance function on the Fluke a good small signal transistor junction measures in the picofarad range; a leaky or shorted junction reads in nanofarads or microfarads. The results were unambiguous.

The entire 2SC984 driver stage: Q11, Q12, Q13, Q14 was failed. Q11 read 1824µF. Q12 read 1857µF. Q13 read 8034µF. Q14 read 84nF. The matched readings on Q11/Q12 and Q13/Q14 respectively suggest they failed in pairs from the same stress events. Q9 and Q10 in the preamp section were also failed.

The complementary PNP stage: Q15 and Q16, original 2SA565 was equally failed. Q15 read 1565µF then open. Q16 read 4265µF then open.

This cascade of failed transistors in the driver stage was the primary cause of the 200mV DC offset measured during bring-up and the reason that 200mV could not be trimmed out with the bias adjustment pot. The trim pot has a limited range. It cannot compensate for a completely failed driver stage.

Every signal and driver transistor on the UA6001J1 board was replaced: All 2SC458, 2SC984, 2SC715, 2SC734, 2SC971 positions → 2SC2240 NPN All 2SA565 positions → 2SA970 PNP

After replacement, DC offset dropped to under 20mV per channel and trimmed cleanly with VRe1 and VRe2, the two 500Ω DC balance pots confirmed from the service manual parts list. The failed driver stage had been the cause all along.

*post transistor replacement, we observed an anomaly. Volume was only apparent and very muted at high volume levels. Something we had replaced on the amplifier board was incorrect but what? We were diligent on confirming polarity but did not test after each replacement as in prior. Turns out we had an overlap of solder between a transistor trace. A quick cleanup and all was restored. ---

The Tuner Board UA5406K3 With the main board complete and DC offset confirmed, the tuner board received its full electrolytic replacement. The tuner board is smaller than the main board and contains fewer electrolytics most of the components here are ceramics, Mylar films, and the various coil and transformer assemblies that handle the RF and IF work. None of those were touched.

One cap on the tuner board required removing a shield to access. The shield is soldered down not a removable lid and came off carefully with heat applied to each tab in sequence. A ton of vintage solder was removed over long working sessions, this was more time consuming than expected. It went back on after the cap was replaced and before any tuner testing began. Running a tuner board with RF shields removed affects the circuit performance and would have made alignment impossible. ---

FM Alignment — The Actual Problem After the full recap, the FM tuner was still unstable. Stations would lock briefly and then drift off. Stereo lock could not be maintained. This was expected, though we did not know it at the outset. Recapping an FM tuner shifts the operating points of the AFC and discriminator circuits. Fresh capacitors with tighter tolerances and lower ESR change the resonant characteristics of the IF and discriminator sections. The factory alignment, calibrated for fifty-four-year-old drifted components is no longer valid for new parts. The factory alignment seals on the trimmer pots were intact. This unit had never been touched since it left the Kenwood factory in 1971. The original factory alignment was simply no longer correct.

Three trimmer pots on the tuner board were adjusted: VRa3 — the AFC adjustment — was the primary culprit. Moving it from the factory position produced immediate and dramatic improvement in station stability. VRa2 — adjusted secondarily for stable reception across the dial. VRa1 — adjusted for correct tuning meter center, 4.0 on the scale. VRa2 and VRa1 were then fine-tuned together until stereo lock was achieved and held. No signal generator. No oscilloscope. Adjusted by ear and meter reference against a known strong local station. Full FM function restored.

A note for future service: the IF transformer slugs Ta3 and Ta5 through Ta8 are original 1971 ferrite. They should not be adjusted without an FM signal generator and oscilloscope. Ferrite slugs this age are brittle — a broken slug inside a sealed transformer is not recoverable, and the transformers themselves are not available. The current alignment is functional. Leave it alone without proper equipment. ---

Potentiometers and Selector Switch Every rotary control received DeoxIT D5 treatment volume, balance, bass, treble, and the input selector switch. The oxidation level on this unit was severe. The balance pot required enough applications and cycling passes to fill an afternoon. The input selector switch had a dead AUX contact that initially appeared to be a wiring fault and turned out to be oxidation severe enough that the signal was arriving at the switch but not passing through.

Repeated application and aggressive cycling, not four passes, forty and eventually cleared every contact. The AUX and tape inputs were restored to full function. All pots went from crackling and intermittent to smooth and silent. The ALPS balance pot was found to be an original ALPS unit still on the board. It cleaned up fully and did not require replacement. When a fifty-four-year-old ALPS pot cleans up rather than needing replacement, you leave it in. ---

The Lamps The fuse-style dial lamps are Stanley S83 type 8V, 0.3A, cylindrical with end caps. Of the positions that required replacement, direct LED equivalents in the same fuse format were sourced. Warm white, drop-in fit. The one original 1971 incandescent that had survived the fifty-four years was replaced along with the others. Running one incandescent among LED replacements means one bulb that will fail again within a few hundred hours. It went with the rest. The dial is now evenly lit across all positions for the first time in what is likely decades. ---

Rear connections

The rear connections heavily corroded were brushed repeatedly with a fiberglass pen until each were clean. This takes careful preparation, the loose corrosion and fiberglass will create a fine dust. Wear proper breathing protection.

The Honest Accounting Twenty dollars for the receiver. Components and supplies,multiple sessions across weeks of work.

The math does not add up to financial rationality and it never did. A KR-3130 in fully restored condition probably exceeds the used market average by a lot. Today an untouched “working” KR-3130 is approximately $175, at best someone cleaned the pots. Retail with a 30 day warranty, big city? It could be $400, vintage gear is desirable. I still doubt it would go through a thorough rebuild.

What it is worth is not being in a landfill. It was built with genuine engineering care. The quasi-complementary output stage, the thermistor bias compensation, the RIAA phono preamplifier, the full AM/FM tuner. It was designed to be serviced and to last. Throwing it away because the capacitors aged out after fifty-four years would have been the wrong outcome. I honed my soldering skills and learned a lot.

It now works correctly across every function. FM holds stereo lock. The dial is lit. The pots are quiet. The DC offset is under 20mV. All inputs are functional. It will run for another twenty-five to thirty years on its new capacitors before anyone needs to open it again.

At the end, it sounds beautiful. Vintage audio math. ---

Notes for Future Service IF transformer slugs Ta3, Ta5–Ta8 — original 1971 ferrite. Do not adjust without FM signal generator and oscilloscope. Slugs are brittle and irreplaceable. Discriminator diodes Det1 — original 1971 germanium devices, left in place. Correct replacement is 1N60 only — not 1N4148. Silicon will not function correctly in this circuit. FM alignment — a proper bench alignment with signal generator equipment would improve weak station sensitivity and stereo separation beyond what was achieved by ear. The unit is fully functional as restored. A future service with proper equipment would bring it to factory specification on fringe reception. VRe1 and VRe2 — DC balance pots, 500Ω. These are the correct offset adjustment pots per the service manual. VRe3 is a separate control and should not be used for DC offset adjustment.

transfertunes.com · Vintage Audio Restoration