This little Harman/Kardon Control Amplifier, model A-401, just came through the shop for a full overhaul. It was in a pretty sorry state to begin with – one channel distorted, one barely gasping for air, and a world of scratchy controls in general need of attention. Despite the electrical shape, though, it was in great cosmetic condition and definitely worth a full rebuild.
Physically, it’s a very unassuming piece of gear with elegantly understated styling, the height of ’70s simplicity. Featuring up to 20W output per channel with low distortion, it’s a great small desktop amp or perfect for pairing with a vintage hi-fi system and will fit in with nearly any decor. No wonder the owner wanted it fixed up!
The inside was untouched, which is always nice to see. It’s a very open, accessible construction design. All of the PCBs could be accessed by removing the top and bottom covers without removing any boards from the chassis itself, and every component was easily accessible. After 40+ years, the original 2SC1030 transistors had tarnished pretty significantly but otherwise everything was in decent cosmetic shape.
For whatever reason, this amp had two tantalum electrolytic capacitors – the blue components on the right of the vertical board – but the rest were all standard aluminum electrolytics. I set out for the component replacement which was very straightforward. As expected, the caps were failing and had started to leave some residue on the boards.
All the electrolytic and tantalum capacitors in this amp were replaced with Nichicon Fine Gold audiophile grade electrolytic capacitors. These are much better components than anything that was available at the time and really help to bring out the best in these vintage amps.
Finding drop-in replacement capacitors is pretty tough, but I got lucky with these. Left, the original 6800 uF 35V capacitors, and right, a pin-compatible drop-in replacement of 6800 uF at 100V. Easy! Running massively de-rated like it is, and with the advantages of modern construction techniques, these main filters will almost certainly last decades. It’s always possible to make modifications to use whatever capacitors are available, and sometimes it’s unavoidable (10 mm-spacing screw terminal radial capacitors come to mind) but it’s much easier, and cleaner, when there’s a drop-in replacement available. This amp could take up to about a 10000 uF capacitor in that position and still be within tolerances, so there’s some flexibility.
Even after all these repairs, it sounded fantastic – from one channel. The other channel was cutting out intermittently and had hum and distortion. While probing, the amp blew both its power supply fuses. It turns out that one of the mica insulators under the output transistors was damaged and wasn’t insulating very well; there was a conductive path to the heat sink. These transistors were pretty worn, anyway, so the owner approved a minor upgrade to a modern, new manufacture TO-3 output transistor with much improved ratings – the MJ15003.
During the troubleshooting process, I did identify the defective 2SC945 transistor in that channel’s driver circuit which was causing some distortion. It was a part of a push-pull pair, but only delivered about 1/10th the gain as its compliment on the other side, thus causing the distortion. Replacing it cleaned up the worst of it, but the channel was still popping and snapping a bit for several minutes after powering on. A quick check of the driver transistors in that channel found a few that were weak, intermittently conducting, and poorly matched – so they got replaced, too.
That took care of the problem!
After replacing the parts, everything checked out! It was time to adjust the bias to ensure it’d be a safe operation, then on to performance characterization.
I was very impressed with the results of the tests. This amplifier was flat +/- 0.3 dB through my distortion analyzer’s signal generator range, 10 Hz – 20 kHz. The MJ15003 is also much more capable than this amplifier is asking of it so there’s quite a lot of headroom, and the more efficient signal path due to the up-rated components plus more powerful output transistors resulted in 50% more measured power output at very low distortion.
At 1 kHz, THD was below 0.1% through most of its original power rating of 20%, beating its specification by 67%. This one, however, was able to deliver 30W with a 150 mV input at maximum volume, still with a THD below 0.5%. That’s audible to many ears but still listenable for the few times you might really need the extra power.
This was a fun project, and it’s going to be a great little amp for a desk or an apartment for a long time.
From the late ’50s, I had this nice example of an EICO HF-81 amplifier come through my shop. These are nice little stereo amplifiers with a built-in phono pre-amp and RIAA equalization and feature about 14W of power output per channel from a pair of push-pull EL84s each at about 1% distortion.
EICO was a popular brand of electronics kit maker back in the day. They made a lot of hobbyist or entry level shop test equipment (oscilloscopes, generators and testers) but had a handful of hi-fi stereo kits as hi-fi listening grew more and more popular. The fact they were assembled as kits makes them pretty straightforward to service, too – the layout is clean and there’s a lot of room on the chassis. It had been worked on a few times in the past – several coupling capacitors were replaced with early film caps in what looks to be the late ’60s/early ’70s. Just recently, the owner reported it was giving noise from one channel and took it to another local shop where the technician replaced a 50 uF capacitor (orange, bottom left) with a 100 uF cap that looks to me like it was probably an old-stock capacitor (blue, bottom left.) They’d clipped the old cap out of circuit but left its shell intact; not exactly what I’d call a best practice. More problems started cropping up shortly thereafter as the rest of the original capacitors degraded, so it came to me for an overhaul. After replacing the parts, there was still noise, distortion, and lower volume on one channel seen on the oscilloscope screen in yellow. Tracing through the circuit, it turns out the CRL couplets in the tone network were failing.
I fabricated new PCBs with the same values to replace the damaged parts and mounted them to the shield.
That fixed the problem right up! After some final tweaks and a burn-in test, she’s ready to go home. Lots of parts out of this one:
It’s going to sound great for years to come!
I had the distinct pleasure of working on one of the earliest Bose® Model 901 Active Equalizers from Series I’s first production run. Serial #230, in fact! It’s made entirely by hand and shows traces of hand re-work as the design was being tweaked – really a piece of history!
This particular equalizer was a gift from Dr. Amir Bose himself to its current and only owner as an undergrad on a factory tour in the early ’70s, and after a lifetime of enjoyment it was time for a rebuild.
Incredibly, this one still has the “Acoustic Suspension Loudspeaker” decal in the center of the treble contour knob.
The owner reported it was having several issues, and opening it up one problem was obvious: the bass contour inductors had broken free of their mounts and were rattling around inside.
It was fairly similar to the later Series 1 Early boards I’d serviced, but had different style inductors and a few extra jumper wires.
It did have discrete Zener diodes installed instead of the reverse-connected transistors. And as for the transistors – the early production models used the 2N3393, and on this one they were so worn out they were only delivering about 20% of their normal gain tested out of circuit. That’s definitely no good!
It cleaned up very nicely, though:
This equalizer is restored back to full performance for its owner to enjoy for another few decades. With all new Nichicon Fine Gold electrolytic capacitors, 1% resistors, new miniature wirewound inductors, rectifier diodes, transistors, Zeners, and a thorough switch cleaning this one is good as new and will sound great for a long time to come.
I just finished a new project over at Rain City Audio – 1934 RCA Victor 119 Antique Radio Repair
An example of the 1939 RCA Model 119 radio made its way through my shop recently. I like RCA’s engineering – they were pioneers of radio and even their entry level models like this set had great performance. The styling is classic mid ’30s, too – great in almost any home….
Here’s another project I just finished at Rain City Audio:
This vacuum tube power supply from an engineer’s test bench can replace a radio or test circuit’s power supply to help with troubleshooting or experiments, with up to 400V for B+ and 6.3V 5A for the heaters. It needed some new capacitors, then fired right up and works perfectly!
I recently got to work on another beautifully reconditioned Westinghouse WR-8 Columaire radio all the way from Bellingham. It was originally in for service with the SPARK Museum, but the repair technician reportedly suffered some health complications and was unable to complete the job so it came down to Seattle to finish it off. I’d previously worked on one of these back in 2013, but this one was in a bit better shape – although it took a long road to get there!
At some point in the past, the radio was badly damaged by an errant candle placed on top which must have fallen over and badly burned the top half of the radio. Afterwards, it was placed in a basement for many years and forgotten about, collecting moisture damage and dust along the way until being brought back out.
The owners spent a lot of time on the cabinet:
and reconing the speaker:
Then it was into the shop for the detailed electrical repairs!
All of the tubes in this radio tested brand new, and all the transformers and coils checked out, so it should be a pretty straightforward repair. Overwhelmingly, the resistors tested in spec also, only a few were slated for replacement. It’s been worked on before at one point, too:
An interstage transformer, normally potted in a top can, was replaced with an open-frame transformer from the late ’30s or ’40s, mounted under the chassis.
This model could be used with a phonograph or a remote control, although neither option was fitted on this model.
I began by mounting new terminal strips under the chassis, soldering them to the lugs on the old capacitor pack.
I took a slightly different approach for the other capacitor pack, using the existing phenolic board as the terminal strip after cutting off the pack capacitor.
For a size comparison, here’s a third capacitor pack containing only a single capacitor – and on top, its replacement, and a quarter for scale.
The owner requested something to make it easier to reset the clock when the time came: Molex connectors! I use 600V connectors with individual connections.
Not the prettiest, but it’ll work. The wire insulation was nicked in a few places, so I reinforced those areas with electrical tape.
Time to fire it up and test—loud and clear! I peaked up the alignment carefully and let it burn in for a while. The lead routing ultimately needed to be tweaked a little bit as there was some interference, but it was easily corrected.
After refitting both chassis and the speaker back into the cabinet, this Westinghouse was ready to go home and make warm, rich, beautiful music for many years to come.
What a beautiful piece! You’d never know how far it had to come to get to this shape. Bringing it back to life so nicely was a joint effort – without the owner’s dedication to the cosmetics of the cabinet, it would never have made it to the electrical stage.
A local client just brought in a very interesting Zenith tabletop radio from just after WW2, the 1946 Zenith 6-D-029 Consoltone mantle radio. This is one of the iconic “boomerang” dials Zenith produced for a couple of years after the war. A variety of sets were made with similar styling; in this case the 6-D-029 is a 6-tube AC/DC radio with a combination of octals and loctal tubes.
This radio was repaired in the ’90s by a radio shop in Portland, OR but came to me in non-working condition. That’s not unreasonable for a radio to need another round of service after 20 years, and it looks like they did good work last time.
Checking out the tubes, other than one which was conveniently labeled (and testing confirmed) “weak”, they all had good filaments and good emissions. Something else was clearly the problem. Given the set’s owners don’t own a tube tester, the tube labeled “weak” must have been from the last time the radio was serviced; this one was the RF amplifier front-end tube.
It’s been pretty well cared for – the cabinet is in solid shape and there’s only a tiny amount of dust inside. Underneath, however…
There are several things going on here. Number one is somewhat obvious: something has released some smoke inside. In addition to that, the electrolytics used in this repair were different ages. There’s the blue CDE dual-section capacitor, and a Jamicon 33uF 450V capacitor hooked up as the second filter. Time to start pulling parts.
That Jamicon capacitor is visibly bulging from one side, and testing confirms it’s definitely dead. Of the CDE dual cap, one section was badly out of spec, and the other section tested open as well.
One of the ceramic disc capacitors blew itself apart – rather violently damaging a mylar capacitor near-by and generally making a mess of things. I replaced all of the same model of capacitor with new 630V film capacitors just to be safe.
After replacing the components it was time for the first power-up. No smoke! But, no sound either. Checking around on the voltages, something wasn’t quite right. B+ on the output tube was about 50% high, and there was no voltage on the screens of any tube.
Careful inspection and wire tracing got to the root of the problem: the wire between two tie points, supplying high voltage to the screens of the output tubes and the plate and screens of the RF and IF tubes, had broken at some point – likely from 68 years of metal fatigue. This was likely the root cause of the original failure: the ceramic disc capacitors were rated at 100V. With 4/6 tubes not conducting, the power supply was delivering over 160V when the nominal operating B+ was about 100V under load. This is well over the rating of those ceramic capacitors and could have caused the violent failure which took the radio out of service.
That’s more like it! A jumper confirms the issues has been located. The new capacitors are rated 630V, so should have no trouble with voltage spikes.
I soldered a 2″ segment of wire to make the connection permanently.
With that repair made, the radio fired up loud and clear!
Time to adjust the alignment a bit:
Then finally, back into the cabinet and ready to go home.
With all new film and electrolytic capacitors, and a replacement tube, this Zenith table radio is going to last a very long time before it needs to be serviced again and sound great the whole time.