The Zenith Model 9S30 from 1936 is a large 9-tube AC superhet circuit radio that receives the broadcast band and two short wave bands. The radio had been serviced many times in its past, but many of the original parts were still in place. I decided to try to reverse all prior servicing and to restore the original chassis appearance to the extent possible. No recent restoration work had been done based on the vintage of replacement components used.
The schematic and a parts list for the Zenith 9S30 can be found on-line on Nostalgia Air.
My antique radio restoration logs
The radio was purchased on eBay and was sold as not working. The seller said that it did not power up. As found, the rectifier tube socket keyway was damaged slightly which allowed the tube (a 5U4G was installed!) to be inserted incorrectly - which probably saved the radio. Its original cabinet finish was in excellent condition,. The original grille cloth was torn. The original knobs were all present. The larger outer tuning knob was cracked. The radio survived being shipped across the country without damage, even though none of the chassis bolts were present! Fortunately it was well packed and was also shipped FEDEX. There was some damage (missing wood) to the bottom front and lower right side of the cabinet. But this damage likely happened prior to shipment, since no missing wood parts were found in the package.
One anomaly noted was that the Antenna-Ground terminal strip had been installed upside down at the Zenith factory. The rivets retaining the strip were original and undisturbed. This meant that the labeling (A or G) was not only upside down, but that the screw terminals functions were incorrect. The antenna wire had to go to the G terminal and the ground to the A terminal in order for the radio to function properly. This must have been confusing to the original owners of the radio! I left the radio the way it was found for historical reasons.
I always attempt to avoid purchasing radios that have been "restored" by radio collectors or flippers, and am looking for either all original examples or those which have been "lightly serviced" in the distant past by radio service shops, rather than peppered with new film capacitors. A service shop would typically replace only enough parts to restore operation, and no more! This radio had been serviced multiple times. There were four service tags attached to the line cord - all with different numbers (claim numbers?). All the tags were from Art's Radio & TV Service in Santa Rosa, CA. The telephone number on the tag was only 4-digits. On the rear chassis was a Sears Service sticker also from Santa Rosa, CA (phone 5000). The radio was shipped by the eBay seller from Petaluma, CA - which is in the same area of the state. All repairs were older - no new parts (such as yellow or orange drop film capacitors) were used. It may be possible to find some information such as the date codes of replacement parts such as filter capacitors. Of course the replacement parts may be much older than the repair, since they may have been in the shop's stock for some time before use.
As parts were removed for replacement, it was discovered that some original parts had been cut free from one of their terminals then tacked soldered back. I assumed that a serviceman did this in order to test the part. And when it was tested OK, it was soldered back in place.
The chassis was very dusty, but not rusty. All tubes and tube shields were removed. The dust was blown off, top and bottom, using an air compressor. The dial, dial pan, dial drive mechanism, and tuning capacitor were removed for cleaning and to replace the mounting grommets. The chassis top and sides were then cleaned using GoJo (white) hand cleaner, 000 steel wool, old tooth brushes and other small brushes. The steel wool was kept well away from tube sockets.
The tuning capacitor was cleaned using my old Heathkit ultrasonic cleaner using dilute ammonia. This required multiple steps, since the capacitor would not fit in the cleaner's small cleaning tank. Alter rinsing, the capacitor was further cleaned using soap, water, and old toothbrushes. Before cleaning, the mica trimmer insulators were removed to prevent damage. Before removal, each trimmer was tightened and the number of 1/2 turns to tight noted. This allowed the trimmers to be returned to their approximate original positions after cleaning, drying, and reassembly. The ball bearings were lubricated using Wells CL-200 automotive distributor cam lubricant (a light grease).
In Zenith schematics, all resistors and capacitors having the same value have the same part number call out. So for example, there may be multiple R2's or C4's on the schematic. Before I start work on the chassis I annotate the schematic so that all parts have unique identifiers. I usually add an alphabetic suffix, so that the part numbers are thus R1A, R1B, etc. I then annotate the chassis photo with these unique part numbers with a red felt-tip pen.
The speaker field coil was OK.
The output transformer was OK.
There were some splits in the speaker cone, and the paper material was very brittle.
The driver transformer was OK.
The power transformer was OK. With all tubes removed, 20 volts AC was applied to the primary winding through a Variac and watt meter. The high voltage winding was then checked for balance across the center tap. The voltage was equal within a few tenths of a volt on each side. A transformer with shorted turns will show a difference of more than a volt or so at 20 volts in. I then applied full line voltage. The wattage draw was very low - less than 10 watts. All filament voltages were correct.
All RF and IF transformers were OK, as well as the RF chokes and the wave trap.
Eight small resistors were out of tolerance by more than 20% (some as high as 50%). All were 1/4 watt or 1/2 watt dogbone resistors. One resistor (R5 - the 2nd detector diode load resistor) had been replaced by a newer type similar to the types used by Philco. Its value did not match the schematic in any case.
R9, the five-section wire wound power resistor (Candohm) had several sections that were not in tolerance, although none were open. Also, I noticed that the resistance of several sections changed if a terminal was moved.
The chassis bolts were missing.
The chassis washers were badly deteriorated and were virtually flattened.
The tuning capacitor drive belt was badly deteriorated.
The tuning capacitor mounting grommets were deteriorated.
One pilot lamp socket center terminal had pulled loose from the socket.
One pilot lamp was burned out, and both were #46 vs. the correct #40.
One pilot lamp diffuser was in pieces and the other very fragile due to age and heat damage.
The IF amplifier (6K7G) grid cap lead insulation was worn and the conductor was shorting to the transformer shield can.
I assume that all paper and electrolytic capacitors are leaky and thus should be replaced (I always "restuff" the original components if possible). I do not replace mica capacitors, but may test them in place if possible (usually this requires disconnecting one lead of the capacitor). Since many of the original parts were still in place I decided to try to maintain the original chassis appearance to the extent possible. Normally I would rebuild all original wax-paper capacitors as well as the filter capacitors in their original cases (restuff them). In cases where an original wax-paper capacitor has been replaced, I attempt to find an original type capacitor in order to restore the radio to its original appearance under chassis. I maintain a collection of branded (Zenith, Philco, RCA/GE) as well as generic (Sprague, CD) dud capacitor just for this situation. Fortunately, Zenith provides the part number of the original part used. I was able to find suitable original Zenith duds to replace all the paper capacitors that had been replaced in servicing. All of these, as well as any remaining original parts, would be restuffed with modern film capacitors.
All of the original filter and other electrolytic capacitors had been removed and replaced. In one case, I found an original dud for C20 (Zenith part 22-420, 10mfd/25 volts) in my dud stock. But I had no idea what the original power supply filter capacitors looked like. That triggered an extensive investigation (as well as some chassis forensics!) See Filter Capacitors.
I decided to leave the (good) replacement volume control in place. It would have been nearly impossible to find a good original control.
When I replace a component, I always remove the original part completely from a terminal. Other components connected at the terminal are protected from heat using old medical clamps (hemostats). I frequently find and remove wire stubs from previous component replacement. Excess solder is then removed using a solder sucker in order to expose terminal holes for reattachment of the rebuilt or replaced component.
All original paper/wax capacitors as well as the duds used to replace original parts removed in servicing were restuffed with new film capacitors. Here are the (restuffed) Zenith duds. They all have the correct Zenith part number as shown on the schematic/parts list. The two capacitors on the left are the same vintage as the radio - the ones with the orange cases are likely vintage 1937 or later:
Four of the original capacitors used were Dubilier Cub types, which are more difficult to restuff. Here is the process I use to restuff these types of capacitors. For the remainder of the original capacitors, as well as the dud replacement, here is my restuffing process:
All of the filter capacitors had been replaced - no original parts remained. The Zenith schematic indicates that a single filter capacitor was used (C15, C16, C17) rated at 16mfd, 4mfd, and 2mfd at 450 volts. My normal process in this case would be to find duds for restuffing that looked at least similar to the originals used. I would look for chassis photos at Radio Museum or other sources, such as from members of Antique Radio Forums. I did in fact find several photographs of radios that used Zenith chassis 5903 (9S30, 9S54 and 9S55) at Radio Museum:
9S30: This chassis used only one capacitor, apparently with a cardboard case, mounted with studs or L-brackets on the right (6F6) side. This chassis did NOT have a hole for an additional capacitor on the opposite side (6A8 side) of the chassis. All of the remaining chassis examples had the additional hole in the chassis. I felt that this was likely the original configuration of the chassis when first shipped. Later production chassis likely went with two separate capacitors: C16 and C17 on the 6F6 side, and C15 on the 6A8 side.
9S30: Two capacitors. A round cardboard case capacitor on 6F6 side and small non-original capacitor on the 6A8 side (this would be C15).
9S54: Two capacitors, one on each side. A metal can type on 6A8 side (C15). There was not enough detail to tell what the capacitor on the opposite side looked like.
9S54: Only one capacitor obvious - 6F6 side. It is not clear what type.
9S54: Only one capacitor - may be cardboard, 6F6 side.
9S55: Two capacitors. 6A8 side - metal can. Not clear what type is used on the 6F6 side.
9S55: Two capacitors. 6A8 side metal can, no cardboard cover, screw base with single positive lug and attached negative ground lug. Insulating washers used between chassis and ground lug. 6F6 side: cardboard case mounted with screws (it likely used threaded studs), with wire leads (the leads had been cut flush with chassis).
I also received some information from an Antique Radio Forums member who has restored several 9S30 chassis. One example he owns has a cardboard case capacitor on the 6F6 side (2 x 4mfd: C16 and C17) and a metal can type with a cardboard cover on the opposite side (C15). My 9S30 chassis originally appears to have had two capacitors: one one each side. Based on evidence found on my chassis, and remnants of the original wire leads, my chassis originally had a cardboard case capacitor with two sections on the 6F6 side (C16 and C17), likely two 4mfd /450 volt units with wire leads, mounted with studs and nuts, and a single metal can type wet capacitor on the 6A8 side (C15). The plus and minus wiring to C15 was still in place in my chassis, although it had been cut short when a replacement capacitor was installed. The plus lead (red) ran to the 80 rectifier pin 8. The negative lead (brown) ran to the B- lug of the candohm resistor R9. This is the configuration I planned on restoring if parts could be found.
For C15 (metal can) and found a Zenith 16mfd 450 volt wet capacitor in my dud stocks (22-294 B)! This part was originally used on a Zenith model 12A58. The capacity and voltage matched the schematic (for part C15), and this could even be the correct type original part. It had a single positive lug and a negative lug which is retained by the screw base mounting nut. The part may have had a cardboard cover originally. My process for restuffing these types of capacitors is as follows:
For C16 and C17, in my dud capacitor stock I found a cardboard case dry type capacitor with wire leads, which mounted with threaded studs. It fit on the chassis, but was likely smaller than the original. The case had no markings, which actually worked to my advantage. I restuffed it with two 4.7mfd/450 volt capacitors. New wire leads were used. The cardboard case was painted using silver/aluminum enamel.
For C20 (10mfd/25 volts) I found an original dud in my stocks (Zenith part number 22-420). It was restuffed using a 22mfd/50 volt capacitor. It was quite difficult to clean out the old contents since it was sealed with tar. The original label was damaged when it slipped off of the tube during the process.
Several sections of the metal clad power resistor R9 (Muter Candohm type) were out of tolerance. Also, the resistance of a few sections varied when the terminal lugs were moved. However no sections were open, and the insulation appeared to be in good condition. These resistors are constructed using clamps (holding the terminal lugs) around the resistance element. The clamps/terminals are not fastened to the resistance strip. Over time, the insulator on which the resistance wire is wound shrinks, or the clamps expand. This results in a high and unstable resistance, or even an open section in extreme cases. I have found that if pressure is applied near each terminal lug, often the resistor can be repaired. My preferred method is to remove the resistor from the radio (usually attached using rivets) and exert pressure on both sides of the case using my large bench vice. Alternately, C-clamps can be used. This is the method I used in this case, since I did not want to disturb the wiring or risk damage when drilling out the rivets.
This "repair" method only has a chance of working if the fish paper insulation is in good shape, with no signs of heat damage or deterioration. If this insulation fails, the result is usually a B+ short to the metal case (which is grounded). This could destroy the power transformer or other parts. Most collectors routinely remove the Candohm and replace it with several wire wound resistors mounted on terminal strips, even if the resistor is good! While this is the most reliable method of repair, it does destroy all originality under the chassis. It is also difficult to find replacement wire wound resistors with the correct resistance and wattage. Often several resistors in series or parallel must be used for each section.
After applying clamps at each terminal location (only one side of the resistor was accessible), all sections were in tolerance (some exactly) and all resistances were stable when the terminal lugs were moved.
Nine small resistors were out of tolerance by more than 20% (some as high as 50%). All were 1/4 watt dogbone resistors. One original resistor R5 previously had been replaced by what looked like a Philco type carbon resistor. The schematic called for a 650K resistor. A 470K was installed. It is unusual that this particular resistor would be replaced, since it is in a difficult position to access, and is in parallel with a mica capacitor. I collect NOS as well as used dogbone resistors just for this purpose, and buy all I can find on eBay and at swap meets. In most cases I did not have replacements available that were in tolerance. In these cases I attempted to find a replacement that was the correct size and had the correct measured value (within 20% tolerance) but not the correct markings! I then repaint the resistor with the value required using hobby enamel paint. In the photo below are the repainted resistors:
Some of the original tubes were re-used if good or only slightly weak. The 5U4G was replaced with a 5Y3G. Correct G type tubes were installed for the 6A8 and 6H6GT. The two 6F6G tubes were good, but very different in quality. One measured 1500 and the other measured 2200 (target was 2000). I also noticed that the hum balancing control was all the way to one side. So I substituted two more closely matched 6F6G tubes.
The center terminal lug of one of the pilot lamp sockets had pulled loose due to failure of the rivet. The socket was disassembled by removing the remnants of the rivet, the internal and two external fiber shoulder washers, and the copper center contact spring. Unfortunately the inner fiber insulator was damaged and could not be reused. A replacement was fabricated using a #6 fiber washer that was sanded down to fit inside the screw base of the socket. The rivet was replaced using a 6-32 brass screw and nut. The head of the brass screw was ground down flat and to size so that it could be used to retain the copper inner contact spring. The socket was then reassembled and the original terminal lug reattached using a 6-32 nut.
Zenith as well as other manufacturers used pilot lamp diffusers to protect the plastic dial from the heat of the pilot lamps. It is very common to see Zenith black dials with holes burned in them. In many cases, the 150ma pilot lamp is replaced by a larger wattage 250ma bulb. In my case, one diffuser was in pieces. The other was intact but very fragile due to age and heat. It was broken during removal. Removal required gentle heating with a heat gun. Once removed and somewhat flattened, a rough template was traced onto a piece of thin cardboard (the back of a writing tablet) The tracing was difficult since the original part was in several pieces. The cardboard pattern was then refined and cut out using an Exacto knife fitted with a new #11 blade. This pattern was then used to cut two replacement diffusers from a suitable piece of white plastic sheet. If I recall, the plastic was originally cut from a food carton.
Cleaned, Lubed and Rebuilt Tuner Mech
The cabinet was vacuumed then cleaned using GoJo (white) hand cleaner and 000 steel wool. Nothing else was done to it. The torn grille cloth was left in place, since it was original.
After completion of the electronic restoration, power was applied through a fused Variac and watt meter. A DVM monitored the B+ voltage. Power was slowly increased while B+ and wattage draw was monitored. The radio came alive and worked on all bands. The only glitches were the tone control (scratchy) and an intermittent tube socket (6C5 first audio amplifier). The tone control was given a shot of GC Big Bath cleaner, after which it worked OK. Pin 5 of the 6C5 first audio amplifier tube socket was not making a firm connection, resulting in intermittent operation, hum, and static. The tube was pulled and the socket contacts gently squeezed using needle nose pliers. After that, operation was OK. The radio worked on all bands.
The speaker had a noticeable rattle at even low volume, and especially if the tone control was set to boost the bass response. There were several bad splits, and the cone material was very brittle. The splits were repaired with several coats of GC Service Cement. After that, most of the rattle was gone.
The radio was then aligned. The instructions given in Riders (Nostalgia Air) are very brief, hard to read, and are incomplete. I received some help from the good folks at Antique Radio Forums and was able to complete the alignment process.
My original restoration objective was to return the radio to its un-serviced condition - to reverse any prior servicing - and to restore operation. Ideally no repairs would be visible. Not all restoration objectives were met, as some of the original replacement parts were not available or were unknown. Some of my "misses" were:
Chassis After Restoration