Sparton Model 10 Radio Restoration

The Sparton Model 10 is an early 7-tube AC non-AVC superheterodyne radio, broadcast band only.  The set features a double-tuned pre-selector (four section tuning capacitor) with RF amplifier stage.  It uses 2.5 volt tubes typical of the mid 1930's.

As acquired, the cabinet was in excellent condition but the grille cloth was torn.  The knobs were original.  There was some loss of finish area at the bottom of the cabinet.  The veneer on the front was in excellent shape (the remainder of the cabinet is solid wood).

The radio had seen some servicing in the past since all the tubes were not the unusual types called for (435, 427, 480 etc.) and were all other brands.  The top and bottom of the chassis appeared all original - no parts had been replaced.  This being the case, I decided to try and retain the original top and bottom chassis appearance if possible. 

The schematic for the radio can be found on Nostalgia Air but it is hard to read and there are several errors.

My antique radio restoration logs

Survey

My usual restoration procedure is to first make a complete survey of the condition of all components.  The survey results guide my restoration strategy.  If major and unique components are defective and cannot be restored, I may elect to sell the radio rather than restore it.  I assume that all paper and electrolytic capacitors are leaky and thus should be replaced (I always "restuff" the original containers if possible).

Repairs

This is a very difficult radio to work on - one of the worst I have encountered.  The chassis is packed full, and components must be removed in order to access components below them.  I had the remove the combination first detector and oscillator coil (inside the copper shield under the chassis - see photo below), both bypass capacitor blocks and the tone control in order to access and test components below them.  

Detector Plate Choke

The detector plate choke was open.  I first tried unwinding some wire from one end, hoping to find a break.  After unwinding about 1/4 of the wire, I gave up.  I cut off the existing wire and rewound the choke using #40 magnet wire by chucking one end in my Unimat lathe running at the slowest speed.  The restored choke was then dipped in rosin to approximate the original appearance.

Paper-Wax Capacitors

The radio contained only three tubular paper-wax type capacitors.  Most capacitors in this radio were either mica or contained in two metal containers.  The tubular paper-wax capacitors were made by Cornell (later to become Cornell-Dubilier?) and branded "Cub".  They were quite unique in construction and could not be re-stuffed in the conventional manner.  Here is the procedure I used to restore them.  The results are shown below.

The capacitor on the bottom is an original Cub capacitor.  The capacitor on top is a reproduction using a new film capacitor, while retaining the original label and end caps.

Bypass Capacitor Blocks

There were two bypass capacitor blocks in this radio.  One contained three 0.5 mfd capacitors (one high voltage and two low voltage).  The other contained seven capacitors or either 0.2 or 0.3 mfd.  Two were high voltage and the remainder low voltage cathode bypass capacitors.  The cans were opened by removing a cardboard cover and the contents removed mechanically.  The capacitor block containing three 0.5 mfd capacitors was rebuilt using three 0.47 mfd 400 volt radial film capacitors.  The original wire leads were re-used in this case.  The other capacitor was rebuilt using six 0.47 mfd 400 volt radial film capacitors and one 0.47 mfd/630 volt film capacitor (this one is across full B+).  New wire leads were used in this case.  This capacitor was filled with rosin to stabilize the contents and the cardboard covers reinstalled.

Mershon Filter Capacitor

The Mershon filter capacitor was two sections: 8 and 16 mfd (voltage unknown).  The main part of the capacitor was below the chassis, secured by a clamp.  The terminals on top were covered by a clamped-on shield.  The capacitor was removed from the set.  I wanted to retain the original terminals and appearance as much as possible.  I marked the area on the nickel plated copper can where the mounting clamp was located, then cut the can in two pieces using a hobby razor saw.  The contents were then removed and the parts cleaned.  Lugs were attached under the original aluminum terminal bolts and new 450 volt electrolytics were mounted.  I used 10 mfd for the 8 mfd section and 22 mfd for the 16 mfd section.  The grounds were soldered to the inside of the top part of the can.  The two halves were then rejoined using the original clamp and secured in a couple of places using solder.  The clamp covers the sawed joint and the solder is barely visible.

IF Transformer

The copper shield was removed from the top, revealing the coil and trimmers.  The top coil (primary) was open.  Luckily the break was right where one end of the coil attached to the terminal (usual failure point).  I was able to scrape the remaining wire clean and form a splice from one piece of wire from a piece of stranded wire.  The splice was wrapped around the terminal and soldered.  The other end was then laid over the stub, the stub folder over, and the wires soldered.  This established continuity.

Resistors

There were five resistors that needed to be replaced.  They were the old style "dog bone" types.  But these were unlike any that I had seen before: instead of a color dot for a multiplier, these used a color band.  I found some dog bone resistors in my junk box and NOS dog bone stock that had drifted to close to the resistances needed. I repainted them using enamel purchased from a hobby store. I assume that the resistors took 60+ years to drift by 40-60%, so future drift would not be excessive (at least in my lifetime).  To me, maintaining the original look is more important than long term reliability of the radio.

There was a wire wound voltage divider resistor that had one section open (the 5500 ohm lower section).  From the voltage table, I calculated that this section dissipated about 2.5 watts.  The replacement would have to be about 12000 ohms 10 watts adjustable.  I could not find anything like that available today (10K and 25K was available).  So I decided to simply shunt the open section with a 6K 5 watt wire wound resistor that just happened to measure 5.6K.  The repair is hardly visible, although I would have preferred to replace the entire resistor.  My fear is that the open section could become intermittent at some point.

The adjustable hum control pot showed intermittent operation.  Cleaning the slider and resistance element did not help. I determined that the problem was the riveted connection between the slider and frame/contact.  The pot was repaired by soldering a piece of fine wire between the slider and terminal.  I formed a small coil in the wire for slack needed for adjustment and to avoid strain on the wire.  This restored normal operation of the control.

Other Repairs

Chassis

The chassis and top components were cleaned using GoJo, steel wool, and toothbrushes.

Testing

After completion of restoration, the radio was connected through a watt meter to a variac and the power gradually applied while monitoring the B+.  A 50' indoor antenna was connected.  The radio immediately came alive and worked well.  The radio was then aligned.  As found, the set was way out of alignment.  In order to adjust the IF transformers, the shields had to be removed - there is no way to adjust the trimmers with the shields in place. The radio works very well with very good tone and sensitivity for an early superhet.  

Post Restoration Photos

Restored Chassis.  The large copper cylinder is the combined mixer and oscillator coil shield.  The two metal boxes on the right are bypass capacitor blocks.  The cylinder on the left is the Mershon filter capacitor.

Below: Front view of restored chassis.  The dial mechanism is quite unique in that the pilot lamp rotates behind the dial as the radio is tuned.  The dial pointer also rotates in front of the dial.

Below: Back view of restored radio.