Restoring a Silvertone Model 6074 Tabletop AC/Battery Set Gerry O'Hara VE7GUH Sep 2011

	Restored Silvertone 6074.

Many radios sit around in folks' garages, attics and basements for years - waiting for either the dump, a moving or garage sale, or, in some cases, a new lease of life. This particular set had done just that - though this time it had been purchased some 30 years prior by the current owner, always intending to return it to working condition and refurbish the cabinet. However, weeks turned to months, months to years and years to decades. So, after some 30 years languishing in this SPARC customers attic, the owner decided that it was 'now or never' and, after deciding it was somewhat more complicated than he could tackle on his own, he brought the set to the SPARC museum to see what could be done.

This is rather a large table model with an attractive 'waterfall' cabinet style, dating from the 1938/9 model year. It is a higher-end AC/Battery model that can be operated from either 115vAC or 6vDC supplies at the flick of a switch. The circuit has a somewhat unusual tube compliment, comprising: a 6D8G converter tube, a #15 as the IF amplifier (465kHz), a 6T7G Detector/AGC/1st audio, a 6L5G audio driver and a 1J6G dual triode push-pull output tube. The power supply switches a transformer between the AC line supply and a vibrator, with a 6ZY5G rectifier in the HT line. Two of the tubes (#15 and 1J6G) have 2v filaments that are provided for by a network of resistors from the 6v AC tapping on the power transformer for AC operation, or from the 6v DC supply on battery operation. The reason for the odd tube line-up appears to be that of battery economy - the tubes are all low-current draw types - for when the set is used on battery power. A standard 4-pin positive ground vibrator is used in a conventional arrangement with hash suppressor chokes and capacitors, all housed in an enclosed 'box' steel sub-chassis along with the rectifier tube and transformer. The well-screened power supply is mechanically isolated from the main chassis on rubber mounts, grounded by a separate braided strap.

Chassis as received.

Dismantling unit.

On receipt, the radio was in rather poor shape: the cabinet was in need of a complete strip-down and re-finish and the chassis was corroded and in a dirty condition - typical of sets with a mechanical pre-set station selector mechanism mounted above the chassis: the lubricating grease on these units (often applied liberally over the years by well-intentioned owners) is a magnet for dust and fluff. This eventually clogs the mechanism, along with any rubber mouldings used for cushioning (usually these have perished/decomposed with age and exposure to lubricants),and in this case it was just a filthy congealed mess. In addition, the friction drive rubber was completed perished, the dial cord was broken and the pointer was missing - all-in-all, a bit of a challenge on the mechanical side. Electrically, all the tubes were present and the underside of the chassis looked to be all-original, dominated by a large multiple (4) cardboard case electrolytic unit. However, a hole was present in the top of the chassis that could have been home to an electrolytic can at some point, however, the chassis layout diagram in Riders did not identify any component in this position (and the hole looked original) - a bit of a puzzle. The speaker - unusually a permanent magnet type (one of the first Alinco magnets?) - sported a large hole in the cone. Most of the other capacitors were Aerovox waxed paper types and all tested leaky/low capacitance. Spot checks on a few resistors indicated that they were within tolerance.

Power supply before restoration ...		... and after.
Chassis below before restoration ...		... and after.

The first job was to render the power supply operational. To gain access to the capacitors, the unit has to be removed from the chassis. At first sight this looks straightforward, however, the wires connecting it to the main chassis had been cut (needlessly) to just the right length during manufacture and thus the only way to remove the unit was to cut (and label) the wires - not the most elegant form of construction. The insulation on these wires (cloth) was frayed where they entered the power supply box and so I decided to replace all the wires to this unit and make them sufficiently long that the unit could be removed in future without disconnecting the wires. To assist in this, an additional tagstrip was installed under the chassis to shorten the flying lead lengths required. Once removed and re-wired to the main chassis, the power supply was easily re-capped (replacing paper capacitors with poly-film types plus one high-voltage disc ceramic) and re-installed. I also installed new electrolytics into the main chassis, including the two power supply filter capacitors. To test the power supply, the 6ZY5G rectifier was installed (all other tubes removed), the set switched to AC operation and powered-up slowly using a variac. The filament supply came up ok, but no HT - the rectifier tube filament was burned-out. As a temporary measure (not having a spare rectifier of this type to hand), I installed two 1N4007 silicon diodes into an Octal tube base and inserted into the rectifier tube socket. Up came the HT voltage, but rather too high - I kept the supply voltage to around 105vAC, this keeping the HT voltage to around 175v DC.

With the power supply now functioning, I replaced all the paper capacitors in the main chassis with poly-film types (630vw), re-installed the remaining tubes, plugged in the speaker and powered-up the chassis using the variac. Stations could just be heard, but were weak and distorted, so something was not right - probably in the audio stages. I signal-traced the audio using an oscilloscope and everything seemed fine up to the grids of the output tube. I located a replacement and in came the audio at good volume and no noticeable distortion. By this time I had located a 6X5 rectifier tube (a reasonable substitute for the 6ZY5 rectifier, albeit with a higher filament current draw). Installing this brought the HT voltage down and with 115v AC applied to the set this was now around 165v DC. I noticed that the filament voltages were a tad on the high side though, but within the 15% or so that is ok (7v and 2.3v for the 6v and 2v filament tubes respectively). Installing four new dial lights dropped the filament voltage slightly. I noticed that the 'Magic-Eye' ('eye') tube was not functioning (probably not surprising as it looked to be an original Silvertone part) - not having a 6AB5 handy, a 6E5 was installed (an acceptable substitute, but with somewhat different characteristics) and found to work reasonably ok, except it was a little too sensitive (this set generates up to -15v AGC on a strong station) - I decided to investigate more later, or try to obtain the correct tube type (eye tubes tend to be expensive, so I contacted the owner prior to purchasing one specially -however, I managed to find a new old stock correct 6AB5 type for a reasonable price). I also noted that the converter tube in the set was a Rogers 6J8M (not the 6D8G per the schematic). The 6J8M is a direct substitute for the 6D8G and so was left in place for the time being as I did not have a 6D8G tube.

Preset-tuning mechanism before ...		... under restoration.
Friction drive before...		... and after.

With the set now functioning reasonably well (on all bands too), I decided to turn my attention to the mechanicals. First order of the day was a thorough clean-up of the pre-set switch unit. A liberal dose of WD40 to start with and wiping much of the greasy grime away. This revealed hardened grease adhering to the switch shafts and cams - this had to be laboriously removed using lighter fluid and Q-Tips. That done, I re-lubed the mechanism with white lithium grease - all working smoothly. Next was the manual dial drive mechanism: this comprises a short (tuning) shaft mounted on a sliding frame that, when pushed-in, engages a chamfered rubber wheel against the dial cord drum, allowing it to be rotated by turning the tuning shaft. The old rubber wheel was perished and useless, so this was removed and replaced with a ¼” chamfered faucet washer, this being fixed to a metal backing washer with superglue and then friction-mounted on the tuning shaft - works a treat. And finally, the moment I had been dreading - re-stringing the tuning mechanism. There appeared to be a missing idler pulley (interestingly the pulleys are made from wood!), but on closer examination and reference to the cord diagram on the schematic, this was not the case. Relieved, I next fabricated a suitable pointer from one obtained from a scrap chassis - this required some cutting to shape with a junior hacksaw and some careful filing of the runner, together with bending of the pointer to suite the scale and slide rail in the chassis. The dial cord is wound around 4 pulleys and two of these turn the cord through 90 degrees as the drum is at 90 degrees to the scale plate. I used .003” dial cord for the job, roughly estimating the length needed, adding around 8” and then cutting this length off the spool of cord. The ends of the cord were sealed (using the heat from my soldering iron) and then carefully threaded around the pulleys - I find that small pieces of masking tape help here to temporarily hold the cord in place on pulleys etc. while threading the remainder through its tortuous route. After an hour or so of this, I had it done and was pleased to find that it worked as expected. The bandchange switch operates a sliding celluloid strip that masks holes in the scale plate to allow a dial light to shine through and indicate which band the set is switched to. The celluloid strip had jammed in its runners and took some coaxing out - however, I managed to work it free, cleaned it up and re-install without too much difficulty, though a new tension spring was needed to return the strip into position reliably.

	Speaker cone repair.

With the mechanicals sorted out, I decided to undertake a full re-alignment (RF and IF) on the set and to set some pre-set stations. First of all, I installed the tube shields (except on the converter tube, as this was a metalized envelope Rogers type). Alignment was straightforward and the set tuned-up nicely (it was not that far out anyway). Finally, I repaired the speaker cone using some stiff paper, model aircraft dope and rubber speaker cement - not perfect, but much better than a hole. Of course, the speaker could be replaced with a modern permanent magnet speaker, however, the original sounds ok.

The only item remaining on the chassis that had not been checked was the vibrator. The old mechanical types (like the one fitted to this set) reputedly only lasted around 18 months in normal use, however, for around $30 a plug-in replacement solid-state one can be obtained that should last indefinitely. Again, a check was made with the set's owner prior to purchasing one of these as the set works just fine on AC and if there is no intention of operating it from a 6v DC supply, there would be no point in going to this expense. I also replaced the low-voltage cables to the set as the old ones were covered in brittle rubber that was flaking off. A solid-state vibrator was obtained, and worked well, however, it was found that if the set was switched to AC supply with this unit installed, the dial lights on the set flickered and the vibrator unit overheated - this condition was traced to a quirk in the sets' circuitry that applied AC to the vibrator when on AC supply through the transformer - this has no effect on a conventional (mechanical) vibrator, but the solid-state unit was suffering some distress and is therefore unsuitable for use in such a set. Amazingly, the (conventional) vibrator that came with the set was found to work fine and so was left in place.

New tubes had been ordered, including the correct rectifier (6ZY5G), audio driver (a 6L5G, as the original tested weak), output (1J6G) and magic eye (6AB5). These were installed in the chassis and found to work very well: particularly satisfying was the 6AB5 magic eye tube which functioned perfectly, showing full deflection on a very strong local signal without the trace-overlap as the 6E5 had exhibited.

View of the refinished cabinet.

Replaced grille cloth.

The chassis repairs completed, I turned my attention to the cabinet: not too bad, but with many scratches - some quite deep - and some damaged veneer along one edge. Restoration of the original finish was not possible, so I started by scraping the lacquer off. This was followed by a repair to the veneer along one edge and some minor filling using epoxy filler, then sanding with 400 then 600 grit wet-and-dry, lubricated with lemon oil. After being allowed to dry for over a week, the sections of the cabinet were masked-off to allow dark-tinted lacquer to be applied to highlight the edges of the cabinet and the speaker grill. The remainder of the cabinet was then tinted with a medium-brown lacquer, with some gradational toning along the edges, followed by many coats of clear semi-gloss lacquer to finish. New speaker cloth was fitted, a gold, diamond-pattern - somewhat brighter, but in keeping with the style of the original style.

The control panel on this model is quite distinctive, being a brown thermo-plastic moulding curved to match the contours of the waterfall front of the cabinet. Unfortunately, this plastic tends to distort with time - particularly if stored in poor conditions (heat, cold, damp, sunlight, etc), and this was the case for this set's front panel - it had apparently been stored for some 30 years in the owners attic. In addition, it was split along one side and the central insert that fits around the push-buttons was split and rather distorted from its original rectangular shape. The only 'fix' for this would either be to substitute this panel for a fabricated reproduction or to glue the panel to a thin rigid backing plate (eg. steel, 'duralumin' or fiberglass-resin) and then trim the plastic to approximate the correct shape. In this case, the moulding was just cleaned-up as much as possible using Novus #2 and #1 and the split glued, however, it was distorted such that it will never look pristine. Still, once the dial glass and knobs were cleaned-up, and the panel/chassis assembly replaced into the cabinet it did not look unacceptable - giving some age-related 'character' and certainly better than when the set arrived at SPARC. Also, it sounds great!