You can E-mail the host of this FAQ
(Trevor Gale) by using this link.
on the Dutch Internet service provider XS4ALL.
The tube complement is not always an accurate guide, except insofar as the presence of a given tube indicates that the set was built after that tube was placed in production. You won't find any 1932 radios using tubes with octal bases or 6.3 volt filament heaters, and you won't find any prewar radios with 7-pin miniature tubes. But you may find a 1946 table radio built to a 1935 design. There are also a few other design features that are very obvious on casual inspection; I'll mention some of them as we go along.
In the following discussion, there are references to the example circuits shown in the RCA Receiving Tube Manual RC-19, dated 1959. This manual is available in reprint from Antique Electronic Supply. Examples 19-1 through 19-4 in particular show examples of four standard circuits that were used, either identically or with minor modifications, in the majority of the smaller "collectible" radios built from the mid-1930's on.
1. The five or six-tube AC-DC radio with 150 ma. tube heaters wired in
series. Example circuit 19-4 shows one of these radios, using 7-pin
miniature tubes. This design is colloquially called the "All-American
Five" by some of us. The design was first built in 1939, using octal
tubes (i.e., 35Z5 and 50L6 in place of 35W4 and 50C5), so it is also
called by some a "35Z5 radio" or a "50L6 radio." I list this design
first, not only because it dominated home entertainment radio production
for over 20 years, but because it is a very simple superheterodyne
circuit. If you study this circuit and know what every component's
function is, and study an example radio of this design, you'll be
prepared to trouble-shoot and repair most post-1935 radios.
These sets do not have a power transformer, and could operate
in places like mid-Manhattan, which had 110 volts DC as its primary
electrical service. Most of these were built as table radios, although
some were installed in small consoles and radio-phonograph combinations.
Virtually all clock radios use this circuit. These are generally
AM-broadcast-only. The tube set shown in the example is one of three
common sets, having either octal, loctal, or 7-pin mechanical design,
but electrically equivalent. Some sets, particularly in the early
postwar period, were built with mixtures of tube mechanical types,
because of tube shortages and availability, and some sets used more than
one configuration during their production runs.
The six-tube version had an RF preamplifier, and was more sensitive than
the five-tube. Example circuit 19-3 shows the same
basic design with an RF preamplifier stage, with tuned output
(three-section tuning capacitor). Many of the six-tube versions used
resistance coupling between the RF preamplifier and the converter stage
(see Diagram no. 3, p. 339, in RC-19, for a resistance-coupled pentode
circuit). The six-tube version was often called a "35L6 radio" because
a 35L6, 35A5, or 35C5 was used, allowing connection of one more 12-volt
heater in the series heater string. In the fifties, some of these radios
were built with a selenium rectifier, omitting the rectifier tube.
Also, a few manufacturers built a four-tube version that omitted any IF
amplification.
Several low-end "boatanchor" communications sets used this circuit,
adding multiple tuning coils and provisions for a beat-frequency
oscillator. Notable examples are the Hallicrafters S-38, S-41, S-119,
S-120, and Ecophone EC-1 series; and the National NC-46 and SW-54.
The tube complements are:
a. First version, built primarily 1938-40.
(note: this design is similar to the 19-4 example, but is its immediate
prececessor, so has a few substantial differences, noted below).
12A8 RF-converter, 12K7 IF amplifier, 12Q7 detector-audio, 35L6 power
output, and 35Z5 rectifier. The first three tubes had small top caps
for the signal grid connections, with either metal or glass envelopes.
The original glass tubes had a "G" suffix, indicating use of an ST-12
stepped bulb envelope. The major difference between this design and
that shown in example 19-4 is the use of a 12A8, which uses a slightly
different oscillator circuit than the 12SA7, 14Q7, or 12BE6. The other
top-cap tubes are very similar to the single-ended octal tubes which
followed, varying primarily in mechanical construction. 12J8 and 12K8
were sometimes used as converters as well. RC-19 unfortunately omits
any circuits for these converter tubes. This version uses a series
resistor in the heater circuit because the heater voltages do not add up
to "near 120"). The proper place for this resistor, electrically,
is between the rectifier heater and the power amplifier heater.
b. Second version, built 1939-ca. 1960
12SA7 RF-converter, 12SK7 IF amplifier, 12SQ7 detector-audio, 50L6 power
output, 35Z5 rectifier. This is almost the same radio, but using
single-ended tubes in the first three stages and a power output tube
with a 50-volt heater. The major difference is in use of a 12SA7 in
place of the 12A8---these tubes are different internally. Note that the
sum of the nominal heater voltages adds up to 122.8 volts, allowing
operation without need for any series resistor in the heater circuit.
c. Postwar version, 1945-mid '60's
12BE6 RF-converter, 12BA6 IF amplifier, 12AT6 detector-audio, 50B5 power
output, 35W4 rectifier. The only difference from b., above,is the use of
seven-pin miniature tubes. All are electrically identical to the octal
versions above. Some sets were built using a mix of seven-pin miniature
and octal tubes, however, the presence of seven-pin miniature tubes
indicates that the set is postwar production.
d. Loctal tube version, 1940-ca. 1960
14Q7 RF-converter, 14A7 IF, 14X7 detector-audio, 50C5 power output, 35Y4
rectifier. Once again, the same radio as version b., using loctal-base
tubes in place of octal. Philco and GE were fond of using loctal tubes.
Note that some radios used a 14B8 converter, which is the same
configuration in a circuit as the 12A8.
The six-tube configuration used the same tube type for both RF
preamplifier and IF amplifier, and the 35 volt heater version of the
output tube. In most cases the RF preamplifier is resistance-coupled to
the RF-converter stage, and the radio used a two-stage tuning capacitor.
Some later versions used movable slug tuning in place of a variable
capacitor. This variation began around 1947, and became more common
during the next decade.
2. Five or six tube AC-DC transformerless radios using 300 ma heaters
wired in series.
These radios were the precursors of the 150 ma. series heater
radios. Some of these radios also included a tuning eye indicator,
typically a 6E5. Total voltage drop of the series heater string was
68-74-82 volts requiring an external voltage dropping resistor of
some sort. These radios often used "ballast" tubes or resistance wire
in the line cord for this purpose.
a. Version using large-base 5, 6, or 7-pin tubes, 1935-50.
6A7 RF-converter, 78 or 6D6 IF, 75 detector-audio, 43 power
output, 25Z5 rectifier. Most of these sets were built before 1938,
although a few manufacturers built them in the early postwar era.
There are more variations on this design than on the 150 ma. heater
designs described above. As noted, some sets had 6E5 tuning eye tubes.
Sets with shortwave often had a 76 triode as a separate local oscillator
for the 6A7.
b. Version using top-cap octal tubes, 1936-1950's.
6A8 RF-converter, 6K7 IF, 6Q7 detector-audio, 25A6 or 25L6
audio, 25Z6 rectifier. This reflects the switch to octal tubes in 1936.
The first three tubes had small top caps for signal grid connection.
The 25A6 is an octal version of the 43; the 25L6 is a 25 volt heater
beam power tube identical, except for heater, to the 35L6 and 50L6. The
25Z5 is a full-wave rectifier (two diode sections), and was usually
connected with the two sections in parallel. However, some
manufacturers, notably Philco, used the two sections to provide voltage
doubling for B+. Radios with voltage doubler power supplies are
AC-only, as a voltage doubler requires alternating current to "pump" the
doubler circuit.
c. Version using single-ended octal tubes, 1939-50's.
6SA7 RF-converter, 6SK7 IF, 6SQ7 detector-audio, 25L6 output, 25Z6
rectifier. Once again, this is a "switch," this time to single-ended
octal tubes. Major circuit difference is in the 6SA7 circuit because of
differences internally between the 6SA7 and 6A8.
This version was generally not built as a "price leader" inexpensive
table radio because of the availabity of 150 ma. tubes that didn't
require a dropping resistor in the heater circuit. It was very often
used as the basis for an upscale AC-DC radio. Some configurations that
you may run across:
1. Shortwave receiver using an additional RF preamplifier,
separate local oscillator, and second IF stage. The 6SK7 was used for
the RF and IF stages, and a 6J5 as a local oscillator.
2. Push-pull audio output, using two 25L6 tubes and a 6J5 as a
phase inverter. This may be combined with the RF-IF additions, above,
and a tuning eye tube (6E5 usually).
Note that use of rectified line voltage gives a relatively low B+, a major limitation in the transformerless design. The primary market for a "full house" receiver that had all of these features would have been the DC service metropolitan areas, particularly New York City, and that is the general area where most "odd-ball" configurations of transformerless sets can be found today. In summary, all of the designs identified in items 1 and 2 above either used the circuit shown in RC-19 example 19-4, or fairly simple variations of the design. There are very few radios with these tube complements that vary markedly from the design, which was established around 1932, and licensed to builders through Hazeltine and RCA patent licenses. In general, the sets that deviate markedly from the standard circuit are a few Philcos and Zeniths, and some off-brand sets that may have been marketed through chain stores with chain store brand names.
3. Postwar AM-FM sets, 1945-up. These were made in two configurations: separate FM front end, and common front end (i.e, RF, IF, mixer, and IF amplifiers. There are many variations on both designs, using 7-pin miniature tubes, loctal tubes, or "hot" octal tubes. The 6SB7Y was a "hot" 6SA7-type tube capable of self-exciting oscillation at FM frequencies, and the 6SG7 a "hot" replacement for the 6SK7. The presence of 88-108 MC FM in a radio always means that it is a postwar set, as this band was not assigned to FM until April, 1945. Manual RC-19 shows an example of an FM tuner in example 19-9. Many AM-FM sets "merged" AM capability into the FM tuner design by using a bandswitch in the RF and converter stages, and by connecting IF transformer coils for 455KC and 10.7 Mc. in series, the idea being that the desired frequency will cause one or the other to resonate (high impedance) and the other will appear as a low DC resistance. The bandswich would also select which IF fed the AM detector, and which detector's output was used to feed the audio section. Example 19-9 also shows both the limiter-discriminator and the ratio detector designs commonly used in FM-capable sets.
This ends the "most common" AC-DC section. Now we will consider
history, and some of the other designs.
Example 19-1 in RC-19 shows a later battery-operated portable, using
7-pin miniature tubes. This design was built after about 1934,
originally using 5-6 pin tubes in ST-12 bulbs; later, octal or loctal
tubes. This circuit also is the basis for most later battery-operated
"farm" sets, some of which were built as floor consoles. Close study of
the circuit will show its resemblance to the 19-4 example. A very
significant difference is the use of filament tubes, and the method of
using a back-bias resistor (R10 in the example) to develop grid bias
voltage for the output tube. Note also that a different local
oscillator circuit is used for the 1R5. This circuit was often used in
the "All American Five" design as well, and is not unique to the battery
design. Resistance values in example 19-1 have been chosen for
operating with a 67.5 volt B battery; otherwise, the circuit is suitable
for operating with a 90 volt B battery.
Example 19-2 shows a typical three-way portable. The term "three-way"
may seem confusing, when the radio can be operated either from the power
line or from batteries. However, the fact that it could operated from
110 volts DC as well as from AC lines was considered noteworthy when DC
domestic service was common; thus "AC or DC or internal battery" are the
"three ways." Note that a modern ricebox radio operating on an internal
battery or with an AC adapter is not "three way" as it will not operate
from a DC line.
Once again, this is the Hazeltine-RCA standard circuit used in examples
19-1 through 19-5, with specific provisions for the three way feature.
Example 19-2 also shows use of a double-tuned RF preamplifier.
Notable are the use of series connection of the receiver filaments,
provision of a rectifier, and a changeover switch. In practise, many
manufacturers provided a dummy line-cord outlet inside the receiver.
Plugging the line cord into this outlet would mechanically actuate the
changeover switch, placing the receiver on battery operation. When
studying this circuit, note in particular the order in which the tube
filaments are wired, and the use of an 1800-ohm resistor (R14) in the
3V4 filament circuit to provide a shunt-feed balance current. The order
of connection of series-wired heaters and filaments is significant in
series-string sets. In this case, the 3V4 is connected to the high end
to provide grid bias for operating, and the shunt resistor provides some
of the plate and screen currents for the tube. The rectifier circuit
shown is typical, although three way portables may use a 35Z5 or a
selenium rectifier. DC output from the rectifier is around 120 volts,
depending on the rectifier used, which requires a large dropping
resistor to feed the receiver filaments. Note the use of two large
electrolytic filter capacitors, C28 and C29, connected to either end of
the 3V4 filament. Small filament tubes require "clean" DC power, thus
these two capacitors filter out both residual ripple from the half-wave
rectifier and audio-frequency variations caused by varying power draw of
the power tube. This circuit arrangement is critical. If any filament
opens, one or both of those capacitors will charge up to the rectifier
output voltage. Also, the design assumes that the rectifier is part of
the voltage-dropping string, and 1.5V filament tubes are limited in
their ability to handle out-of-tolerance filament voltage.
The circuit shown in figure 19-3 for an AC-operated receiver is the same
as that in figure 19-4, with several upscale features, and resistance
values selected for operation at 250 volts B+ rather than 120. Note
that the circuits for the 6BE6 converter, 6BA6 IF, and 6AV6
detector-audio stages have the same configuration as those shown for
those three stages in figure 19-4. An additional 6BA6 RF preamplifier is
provided for higher gain and better selectivity. A pair of 6AQ5 tubes
provides push-pull output. The second 6AV6 placed ahead of the lower
6AQ5 grid circuit inverts the audio signal for grid drive, with
"approximately unity gain," determined by the tapped grid leak
(470K/8200 ohms) in the top 6AQ5 circuit. This particular circuit is a
classic example of older home entertainment engineering, and there is
much to criticize in its selection over the use of a twin-triode
balanced paraphase using a 12AX7 or a 6SN7. Why was it chosen? Habit,
probably---it was a good choice for 1932.
The main feature of this set which differs from AC-DC configuration is,
of course, the use of a power transformer and a 5Y3 full-wave rectifier.
The configuration of the rectifier circuit was one of the earliest and
most durable circuits in the history of tube-type home entertainment
radio. This later configuration uses a 5Y3 instead of an 80, has larger
filter capacitors (20 mfd rather than 8 or 10 mfd), and a resistor in
place of an inductance between the two filter sections. Older radios
most often used a speaker field coil between the two filter sections,
partly because Alnico magnets were not available until the late
thirties, and partly because inductance at this point compensates for
using smaller capacitance values to get good filtering.
Note the configuration of the screen circuit for the 6BE6 and two
6BA6's. All three screens are connected together. This is poor design,
and likely to cause parasitic oscillations. The circuit in figure 19-4
also shows the screens connected together, but in this instance, there
are only two screen, in stages that operate in opposite phase, so any
coupling between the two stages has a negative feedback effect.
Older radios:
Home entertainment radio began in 1920. KDKA in Pittsburgh generally
has gotten credit for being the first commercial broadcast station. The
two major receiving tubes available at the time with the UX201 and the
UV199, as they were called at the time. The UX201, later revised and
called 01A was a low mu triode. The V99, as the UV199 came to be
termed, was derived from a telephone amplifier triode, developed
during WWI. Several manufacturers built sets, but the most predominant
in the collector market is the Atwater Kent neutrodyne TRF set using
01A's driving headphones. A standard inexpensive set used regenerative
feedback to achieve gain. These were prone to oscillate, squawk, and
whistle, and created no end of radio frequency interference, and rapidly
lost favor, particularly in high-density metropolitan areas.
The first commercially significant superheterodyne receiver was the
RCA "catacombs" receiver of 1924. This set used V99's, a 42 KC IF
frequency, and a headphone-driving-a-horn "loudspeaker." Both the
A-K and the RCA sets required three DC voltage supplies.
The A supply (5 volts DC for 01A, 3.3 volts DC for V99) heated the
filaments. The B supply, typically 90 volts, provided plate voltage.
The C supply, ranging between 9 and 15 volts, and connected as a
negative supply, was used to bias the tube grids. RF gain was
controlled by a rheostat which controlled the filament voltage. These
three voltages were supplied by lead-acid storage batteries, with a
Tungar bulb charger for charging the batteries when the radio was not
being used. All of the RF stages, and the catacombs superhet local
oscillator, were tuned by separate dial knobs.
If this sounds like the definition of a kloodge, it was. I had examples
of both an O1A Atwater Kent and an RCA "portable" (ran on dry batteries)
catacombs set, complete with lead-acid batteries and Tungar charger, at
the end of WWII. These sets sold by the thousands, but were obsolete by
1929, and most of them were discarded when their storage batteries wore
out. Worth noting that "Philco" is a contraction of "Philadelphia
Storage Battery Company." It is also worth noting here that RCA, or
"Radio Corporation of America," was not a separate company until 1929,
but a patent pool and sales company owned by General Electric,
Westinghouse, and AT&T. The phonograph fans will, no doubt, describe
how the Victor Talking Machine Company and Radio Corporation of America
became RCA Victor.
Automatic volume control methods were developed around 1925. AVC, which
is synonymous with the term "Automatic Gain Control" (AGC), allowed sets
to operate at much higher input sensitivity, and to reduce that
sensitivity to prevent overloading in the presence of a strong signal.
Methods of tracking RF stages and a local oscillator operating at some
difference frequency were also developed in the mid-late 1920's. The
final developments needed to build a mains-powered single knob tuning
"modern" superheterodyne radio were filaments capable of working on AC
without developing hum, a suitable high-voltage rectifier, and a tube
with high plate resistance. The first two appeared around 1928 in the
form of the 26 and 71A tubes and the 80 rectifier. While these were not
the actual "first" devices, they appear in almost all of the early
mains-powered radios. The third came about a year later in the form of
the UY224 tetrode, later known as the 24A. The 24 also had another
recent innovation, the indirectly-heated cathode, which allowed the
cathode element of each tube to "float" at a different voltage from the
heater supply DC reference.
Problems with secondary emission from the 24 were "cured," more or less,
by processing the plate material to reduce this emission. This produced
the 24A. However, a more permanent fix was to include a third grid to
"suppress" the reverse current resulting when plate voltage was lower
than screen voltage. The 57 and 58 pentodes were the result. Both have
2.5 volt indirectly-heated cathodes. However, the 58 has a
characteristic known as "variable-mu." Actually, with pentodes, one
considers transconductance, and what "variable-mu" actually does is to
reduce the transconductance as the tube is more heavily biased. The
feature is desirable in circuits with AVC. These pentodes showed up
around 1931. The pentode power amplifier was also introduced around the
same time, with the 47 replacing the 45 in many designed of the 1932-34
era.
The last significant development in tube design for AM broadcast radios
was the development of a single tube with two control grids to serve as
a self-exciting local oscillator and mixer amplifier. The 2A7, quickly
replaced by the 6-volt-heater equivalent 6A7, was the predominant
design, and the 6A7 was used very commonly until after 1940. The 6L7
also was introduced fairly early. This is a mixer that is not designed
to operate as a self-oscillator, and was used, particularly in
communications sets, with a separate local oscillator, until the
1950's.
Availability of a single tube for the superheterodyne oscillator-mixer
function was essentially the death-knell for TRF designs. Another
contemporary development which entered production in 1933 was the 2E5
"tuning eye" tube, which varied a shadow area on a visible target as an
inverse function of the control grid voltage. TRF sets were built into
the 1950's, but are not very common. They tend to be either very cheap
radios for use in metropolitan areas with strong signals or in high end
sets where the broad bandpass allowed "high fidelity" (though the
AM stations actually only transmit a signal that has 5KC as the 3db
half-power point in the modulation).
Availability of components for a vibrator power supply made automobile
sets operating from 6 volts DC practical. There was a wholesale switch
from 2.5 volt heaters to 6.3 volt heaters in 1934. The 2.5 volt heater
series of tubes quickly became obsolete. The switch to 6.3 volt 300
ma. filaments was parallelled by development of a two-diode rectifier
and an output tube with 25-volt 300 ma. heaters, making series string
wiring of the heater circuit practical. These are the 300 ma. heater
transformerless sets described above, which date from about 1934.
Octal-based tubes enter the picture in 1936. Many of the original
designs were built in self-shielding steel envelopes. Metal octal tubes
were built with a flat "button" glass seal, which allowed much shorter
electrode lead connections. Early glass octal tubes continued to use
the older "press" design, with relatively long leads. RF and AF tubes
in the original octal series had small top caps for connection to their
control grids. It was not until about 1939 that single-ended tubes
entered production.
Development of a button seal that could be used with glass envelopes
allowed manufacture of metal-based "loctal" tubes. These entered
production in 1939. At the same time, a cylindrical bulb for glass
tubes also entered production, allowing closer spacing between tubes.
Experimental FM became a commercial broadcast enterprise in 1940. The
original FM band began at 42 megacycles, and production of home
entertainment receivers to receive that band began in 1941. The band
originally overlapped the experimental television band (later channel 1,
48-54 megacycles). The FM band was reallocated to 88-108 megacycles in
the spring of 1945, thus a set with 88-108 capability is postwar.
Another "strictly postwar" feature is the 7-pin miniature tube. The
9-pin miniature followed around 1949.
A few tubes were "survivors" through the 1928-50 period. The standout
among these is the 80 rectifier, which was still being used in new
production in the mid-1950's. The 5Y3GT which replaced it is nothing but
an octal-based version of the 80. The 2A3 and 45 power triodes, as well
as the less-common 6A3 were all used from the early 1930's until well
into the 1950's. There remains today something of a cult that
believes that these triodes are the only audio power tubes worth
considering. All of these tubes use filament cathodes, and the most
practical circuits for using them required a separate filament winding,
elevated to the 40-60 volts needed to bias these tubes near cutoff.
You can E-mail the host of this FAQ
(Trevor Gale) by using this link.
on the Dutch Internet service provider XS4ALL.
Click on this link to go back to the FAQ index.