One of Philco's principal innovations to their product line for 1937 was Automatic Tuning in combination with Magnetic Tuning, initially
offered on three high-end floor models. The mechanism was covered by US patent #
2,248,678, filed on Aug 6th 1936, with Glaser,
Bowman and Briggs named as inventors. Automatic tuning systems had been available in one form or another since the late 1920s but
the early implementations shared two big drawbacks, namely mechanical imprecision and frequency drift, and as a result they never
really caught on. Philco had been working on station selection systems prior to the development of their 1937 line, as in 1935 they
patented (patent #
2,115,624) a mechanical "fixed channel selecting system" that looked very much like a forerunner to the system they
would eventually adopt. However in the patent they cited the need for a manual fine adjustment capability to allow the operator to
fine-tune the mechanism to account for variations with "
temperature, age and other factors". Perhaps as a result, it was never adopted
for any of their domestic broadcast receivers. What was in fact missing from the early embodiments was a means for automatically
correcting the residual tuning errors, but that all changed for Philco with the introduction of their 1937 line, featuring Automatic-
Frequency-Control (AFC).

Once AFC, referred to by Philco as "magnetic tuning", arrived on the scene in 1936, the limitations inherent in automatic tuning all but
vanished. AFC is an electronic technique designed to "lock" on and "track" a station so as to keep it optimally tuned at all times, relieving
the listener of this chore. By this expedient, audible distortion caused by imprecise tuning was minimized, which was of particular benefit
to the new generation of high-fidelity receivers whose transparency of reproduction rendered this distortion only too apparent. AFC
served as a great "enabler" for automatic tuning and over the next two or three seasons a preponderance of weird and wonderful tuning
contrivances would be offered to the public by numerous manufacturers (see my
Sparton 1268 and Stromberg Carlson 350R pages for
other late 1930s examples of Automatic Tuning with AFC and my  
Stewart Warner 91-513 and Zenith 5-R-317 pages for examples of
receivers without AFC).

Within the space of a few years, nearly all manufacturers, including Philco, would abandon their mechanical systems and standardize
their automatic tuning around simpler, more reliable electrical push-button arrangements that in most cases operated independently of a
set's manual tuning. These systems, at least for AM, could get by without AFC and were much less expensive to produce. As a result of
the cost reductions, by the early 1940s a plethora of small table sets was being offered featuring automatic push-button tuning, in
addition to numerous floor models. But that being said, we radio collectors look back with reverence upon the sets of the mid-to-late
1930s, with their diverse and ingenious mechanisms, for they are some of the most fun-to-use tube radios ever produced!

The Origins of AFC. Systems for automatically adjusting the frequency of oscillators can be traced back to at least 1917. R.V. Hartley
of the Western Electric Company, in US patent number 1,774,003, describes a methodology for synchronizing the frequency of a
receiver's local oscillator (LO) to that of a radio-telephony signal received from a remote transmitter. His circuit provides a discriminator
for detecting the extent and direction of the frequency error between transmitter and receiver, a means for smoothing the resultant
"error signal", together with a means for using that signal to tune the LO such as to minimize the error. In other words, the Hartley circuit
of 1917 comprised all the essential elements of AFC, ones that still form the core of all modern day embodiments.

The Hartley circuit of 1917 used an electromechanical means for adjusting the LO. Based upon my researches, some of the earliest
work using a control tube (later on called a reactance tube) to accomplish this entirely electrically appears to have been undertaken at
RCA circa 1931-1932; US patent numbers
2,065,565 (Crosby) and 2,058,411 (Carlson) describe such approaches. Crosby, in his same
patent, described a discriminator utilizing phase changes in the received signal, as it moves off the point of correct tuning, as a means of
generating the control (error) signal for the reactance tube. Prior to that point, a pair of resonant circuits, one tuned slightly high and the
other slightly low, feeding differentially connected detectors, a la Hartley, had been used. The phase-change approach was considerably
more stable and would ultimately be refined into its most commonly adopted form by another RCA employee,
Stewart Seeley (see his
1935 patent, #
2,121,103). The reactance tube and phase-detector circuits became almost universally adopted for AFC circuits going
forward, right through the end of the tube radio era. Furthermore, with the cooperation of D. E. Foster, Seeley's circuit would evolve into
the Foster-Seeley discriminator, one of the best-known traditional methods for achieving frequency demodulation of broadcast FM.

The AFC architecture for Philco's 1937 and 1938 lines appears to have been the brainchild of Charles T. Travis, B.S, Ph.D (1883 -
1941). His US patent number
2,240,428, entitled "Electrical Circuits", filed in July of 1936 and assigned to the Philco Radio and
Television Corporation, covers much of the circuitry. Travis graduated from the University of Pennsylvania in 1902 with a civil engineer-
ing degree. Between 1903 and 1906 he studied under the Harrison Scholarship program at the University, receiving his Ph.D in 1906.
Between 1926 and 1931 he worked for Atwater Kent and from 1931 to 1935 at RCA. Prior to these dates he had been employed at the
Tacony Steel Company in Philadelphia and at the
Rock Island Arsenal in Illinois [1][2]. Sometime during 1935 Philco must have offered
him a sweet deal and enticed him away from RCA to work on the AFC for their 1937 models (introduced in the summer of 1936) and, in
due course, their 1938 sets. Beforehand, whilst at RCA Travis would no doubt have worked with the likes of Carlson, Crosby and Seeley;
the work performed at RCA by this pioneering group formed much of the basis of the AFC and FM demodulation circuits adopted by
numerous manufacturers in the mid-to-late 1930s time-frame and beyond. Certainly Travis was quite a catch for Philco, as their AFC
circuits work extremely well and embody several innovative features aimed at making them reliable for mass production.

Although modern AM/FM broadcast receivers utilize synthesizer tuning which obviates the need for AFC, AFC nevertheless forms an
essential component in numerous modern-day communications systems, where it is used to combat errors in frequency sources as well
as Doppler induced by host dynamics. One example is in the Global Positioning System (GPS) receivers finding such widespread usage
today. Of course, many of the modern systems are implemented using digital electronics and software, but the underlying principles are
unchanged from those pioneered by the highly-talented engineers back in the early days of radio.
Philco Automatic Tuning and Automatic-Frequency-Control
A historical perspective
[1] "Contributors to this Issue", Proc.of the Institute of Radio Engineers, Vol 23, No 10, Oct 1935.
[2] "Charles Travis. Geologist and Radio Engineer Dies in Philadelphia at 58", New York Times, Aug 17th 1941
TubeRadioLand:Home > various > Auto Tuning, etc